The Transkribus Ecosystem

Features

Where Transkribus is used

What Transkribus is used for

Resources

Plans & pricing New

Transkribus Blog

From census records to trade ledgers, a handwritten table in a historical document often holds a wealth of valuable information. Yet much of this data remains locked on paper, accessible only by manually deciphering each page. Imagine the potential if these tables could be transformed into digital, searchable spreadsheets, allowing you to analyse whole volumes at the click of a button.
With AI tools like <a href="https://www.transkribus.org/">Transkribus</a>, digitising tabular data in this way is now a reality. Transkribus enables users to convert handwritten tables into structured, searchable data, unlocking the stories within and making research both more efficient and more effective.
In this post, we will show you how to convert handwritten tables into digital spreadsheets, helping you bring historical records from paper to pixels.
 
<h2>Key takeaways:</h2>
<ul>
 <li>Tables in paper-based documents can be converted into digital spreadsheets using Transkribus.</li>
 <li>This is achieved by <a href="https://www.transkribus.org/table-models">creating an AI model</a>, using it to recognise the tables in your documents, and exporting the data as a spreadsheet file.</li>
 <li>The digital spreadsheet can then be searched for specific data, or published online through <a href="https://www.transkribus.org/sites">Transkribus Sites</a>.</li>
</ul>
 
<img style="width: 100%; height: auto;" src="https://s3.eu-central-1.amazonaws.com/readcoop.cis.public-assets.prod/blog/Screenshot%202024-11-20%20135717.jpg" alt="Transkribus uses the power of AI to turn handwritten tables into digital data.">
 
Transkribus uses the power of AI to turn handwritten tables into digital data. © “Carnegie Corporation Register of Applications from Educational Institutions, 1911-1920” (Carnegie Corporation of New York Records)
 
<h2>Bring your handwritten tables from the archive to the computer</h2>
The tables in historical records provide valuable insights into aspects of life at particular times and places. But no matter if the table is handwritten, printed, or a mixture of both, its potential is often locked within physical documents. By digitising these tables and converting them into a format like Excel, researchers can search, analyse, and reorganise data in ways that simply aren’t possible on paper.
For example, you could transform lists of trades from centuries-old apprenticeship records into a dataset that reveals changing economic trends over time. Or you could convert registers of handwritten addresses into a searchable resource for genealogical research. Once digitised, these tables can also be uploaded to platforms like Transkribus Sites, where they can be shared and accessed worldwide.
 
<h2>What kind of records can be converted into spreadsheets?</h2>
Not all tables are equally easy to digitise. For best results, the tables should have a consistent layout across each page, with similar handwriting styles and clear separation between rows and columns. High-quality scans are essential, as blurry or poorly lit images can make it difficult for AI to accurately identify table boundaries or recognise text. Historical records that meet these criteria, such as census tables or registers with well-defined sections, are ideal candidates for this digitisation process.
 
<img style="width: 100%; height: auto;" src="https://s3.eu-central-1.amazonaws.com/readcoop.cis.public-assets.prod/blog/Screenshot%202024-11-20%20140014.jpg" alt="The columns and rows in a table should be separated by a line or blank space.">
 
The columns and rows in a table should be separated, whether by a line or blank space. © “A catalogue of pictures” (Paul Mellon Centre for Studies in British Art)
 
<h2>How to convert handwritten tables into spreadsheets</h2>
Below is an outline of the workflow for turning handwritten tables into digital spreadsheets. More detailed information about each step of the workflow can be found in our <a href="https://help.transkribus.org">Help Center</a>.
 
<h3>1. Train a Table Model</h3>
The first step is to <a href="https://help.transkribus.org/tables#how-to-train-a-table-model">train an AI Table Model</a> that can locate the columns and rows in your handwritten table. To do this, you first need to prepare training data. Upload scanned images of your document to Transkribus, and use the Editor to mark where all the rows and columns are. You will need to prepare between 20 and 50 documents as training data, depending on the complexity and homogeneity of your tables.
 
<img style="width: 100%; height: auto;" src="https://help.transkribus.org/hs-fs/hubfs/Draw%20columns.gif?width=688&height=387&name=Draw%20columns.gif" alt="Ground Truth training for AI.">
 
Save these images as "<a href="https://www.transkribus.org/blog/what-is-ground-truth">Ground Truth</a>" and use them to train a Table Model. You may need to retrain the model a few times before it is sufficiently accurate.
 
<h3>2. Run the table model</h3>
Now that your table model is ready, you can <a href="https://help.transkribus.org/tables#how-to-use-the-trained-table-model">apply it to the rest of your document collection</a>. This process tells Transkribus to identify the columns and rows in each table in your documents, creating a structured map of the tables.
 
<img style="width: 100%; height: auto;" src="https://help.transkribus.org/hs-fs/hubfs/Table%20model%20recognition%20(1).gif?width=484&height=272&name=Table%20model%20recognition%20(1).gif" alt="Applying the trained table model.">
 
<h3>3. Perform the layout recognition</h3>
Before performing the text recognition, it is important to run a separate layout recognition first. This allows Transkribus to locate the text within each cell of the table.
When running the layout recognition, you should also <a href="https://help.transkribus.org/advanced-layout-configuration-settings">select the following advanced parameters</a>: "Keep existing regions" and "Split lines on region border". These parameters make sure that the table format recognised in Step 2 is not changed through the layout recognition.
 
<h3>4. Perform the text recognition</h3>
Once Transkribus has located the cells and the text contained within them, it can then <a href="https://help.transkribus.org/automatically-transcribing-your-documents">recognise the text</a> and create a digital version of it. This is done in exactly the same way as a "normal" text recognition. A digital version of the table will be displayed on the right side of the Editor screen, and this can be corrected and edited as required.
 
<div style="text-align: center;">
<iframe width="560" height="315" src="https://www.youtube.com/embed/1y3ZABVnRmU" title="YouTube video player" frameborder="0" allow="accelerometer; autoplay; clipboard-write; encrypted-media; gyroscope; picture-in-picture" allowfullscreen></iframe>
</div>

 
<h3>5. Export as an Excel file</h3>
Finally, <a href="https://help.transkribus.org/downloading">download the data</a> to your computer as a spreadsheet file. This allows you to open it in programs such as Microsoft Excel or Google Sheets.
 
<h2>Can I export all the document pages as one spreadsheet?</h2>
Yes, Transkribus allows you to choose between exporting each page as a separate sheet within the spreadsheet or merging all pages into one comprehensive spreadsheet. This flexibility makes it easy to structure your output in a way that suits your research goals.
 
<h2>Can I export it in another digital format?</h2>
Your tabular data doesn't have to be exported as an Excel file. Files can be downloaded from Transkribus in various different formats, from PDF to <a href="https://tei-c.org/">TEI</a>, meaning you can choose the format that allows you to process and analyse your data most effectively after processing it in Transkribus.
 
<img style="width: 100%; height: auto;" src="https://help.transkribus.org/hs-fs/hubfs/Create%20my%20first%20sites%20(1).gif?width=688&height=387&name=Create%20my%20first%20sites%20(1).gif" alt="Various export formats supported by Transkribus.">
 
<h2>How can I improve access to my documents?</h2>
Once your tables are digitised, consider uploading them to <a href="https://www.transkribus.org/sites">Transkribus Sites</a>. This platform allows you to share your digital records online, making them accessible to researchers, historians, and the general public worldwide. By publishing on Transkribus Sites, you enhance the visibility of your work and contribute valuable resources to the wider research community.
 
<h2>Can I use Transkribus to digitise other types of documents?</h2>
Certainly. Transkribus is designed to digitise a wide range of documents, from medieval manuscripts to handwritten notes. It can also be used as <a href="https://www.transkribus.org/en/blog/using-transkribus-for-ocr-with-printed-books">optical character recognition (OCR) software for printed books</a>. Its versatility makes it a valuable tool for any archive, library, or research institution working with historical records.
 
<h2>Where can I get more information on this topic?</h2>
The <a href="https://help.transkribus.org/">Transkribus Help Center</a> has information and tutorials for all the different features and workflows in Transkribus. In the <a href="https://help.transkribus.org/tables">Table Models section</a>, you can find step-by-step instructions for training a Table Model and applying it to your documents.
Alternatively, check out the recording of our <a href="https://www.youtube.com/watch?v=ZXZkjEa45Ew">Table Models webinar</a> on our <a href="https://www.youtube.com/@transkribus">YouTube channel</a> for a walkthrough of how to train a Table Model, and visit our <a href="https://www.transkribus.org/events">Events page</a> for information about upcoming webinars for both beginner and advanced users.
<div style="text-align: center;">
<iframe width="560" height="315" src="https://www.youtube.com/embed/ZXZkjEa45Ew" title="YouTube video player" frameborder="0" allow="accelerometer; autoplay; clipboard-write; encrypted-media; gyroscope; picture-in-picture" allowfullscreen></iframe>
</div>

From paper to pixels: How to convert handwritten tables into digital spreadsheets

Retraining a text recognition model in Transkribus is one of the most effective ways to generate accurate transcriptions. Whether you're working with historical documents, handwritten diaries, or unique typefaces, retraining allows your model to learn from each cycle of corrections, driving down error rates and enhancing reliability.

In this post, we’ll walk you through the retraining process in Transkribus, explaining why retraining is valuable, what steps are involved, and answering common questions about model retraining.

 <h2>Why should you retrain a model?</h2> 

You've trained your first Transkribus model, and it's delivering decent results. But what if you’re aiming for even greater accuracy?

Retraining models is the best way to achieve optimal accuracy. Each time you retrain your model, you add more <a href="https://www.transkribus.org/blog/what-is-ground-truth">Ground Truth data</a> to it, giving the model more material to learn from. And the more your model learns, the more likely it is to produce accurate transcriptions. That’s why retraining is one of the best ways to <a href="https://readcoop.eu/how-to-improve-the-cer-of-your-model/">improve the Character Error Rate (CER)</a> of a model.

However, while we talk about “retraining a model,” the process doesn’t actually involve updating the old model, but rather creating a new version of it. Each time you retrain your model, you will create a whole new model. Users often label these successive models by version number, such as “Wolpi Model v1.0,” “Wolpi Model v2.0,” and so on, making it easy to track each step in the model's evolution.

 <img src="https://s3.eu-central-1.amazonaws.com/readcoop.cis.public-assets.prod/blog/training-models/image%20%286%29.png" alt="Workflow image" style="max-width: 100%; height: auto;"> 

 <h2>What is the workflow for retraining a model?</h2> 

Let’s say you want to transcribe an 800-page handwritten diary in English. You manually transcribed 50 pages, and used these pages as Ground Truth to train a model: Diary Model v1.0. You use Diary Model v1.0 to recognise the other 750 pages and find that while the accuracy is good, you think you can do better. So you decide to retrain it and create a new version.

The retraining workflow can be broken down into the following steps:

 <img src="https://s3.eu-central-1.amazonaws.com/readcoop.cis.public-assets.prod/blog/training-models/Screenshot%202024-11-07%20114508.jpg" alt="Workflow process image" style="max-width: 100%; height: auto;"> 

<ol>
 <li>Recognise more pages. The first step is to use the first version of your model to <a href="https://help.transkribus.org/automatically-transcribing-your-documents">recognise more pages from your collection</a>. In this case, we would use Diary Model v1.0 to recognise another 20 pages from the diary.</li>
 <li>Accurately correct the text. Take the automatic transcriptions generated by the model and <a href="https://help.transkribus.org/transcribing-manually">manually correct them</a> to make sure they are 100% accurate. The more accurate your transcriptions are, the more accurate your model will be.</li>
 <li>Save the pages as Ground Truth. Once your transcriptions are finalised, <a href="https://help.transkribus.org/data-preparation">save them as Ground Truth</a>. In addition to the 50 pages you used to train Diary Model v1.0, you should now have 70 pages of Ground Truth.</li>
 <li>Train a new version of the model. Use those 70 pages of Ground Truth to <a href="https://help.transkribus.org/training-text-models">train a new model</a> — Diary Model v2.0. If there is a relevant public model that fits your collection, then you can use this as a base model. However, do not select the previous version of your model as the base model, as this will confuse Transkribus and produce poorer results.</li>
 <li>Recognise more pages. Once you’ve trained the new and improved version of your model, you’re ready to start the whole process again. Use Diary Model v2.0 to recognise another 20 pages from the diary, manually correct them to create more Ground Truth, and then train a new version of the model with these 90 pages. This process can be repeated as many times as you need until you no longer see any improvements in accuracy.</li>
</ol>

It is important to remember that a model will not improve until you train a new version of it. People who are new to model training sometimes mistakenly believe that correcting the transcriptions will automatically improve the model. However, only by following the steps above and actively training a new version of the model can you increase its accuracy.

 <h3>How long does this process take?</h3> 

Reading the description above, it may seem like retraining a model is quite a time-consuming process. However, bear in mind that each new version of your model should be more accurate than the last, and so each retraining should produce more accurate transcriptions that require less manual correction.

In the example above, it might take a few hours to correct the transcriptions produced by Diary Model v2.0. But by the time you’ve trained Diary Model v5.0, the transcriptions might be so accurate that they barely need any correction at all.

 <iframe width="560" height="315" src="https://www.youtube.com/embed/KVQpM3CpJ4M" title="YouTube video player" frameborder="0" allow="accelerometer; autoplay; clipboard-write; encrypted-media; gyroscope; picture-in-picture" allowfullscreen></iframe> 

 <h3>How many pages of Ground Truth should I add each time?</h3> 

This is a difficult question to answer as it depends on many factors, such as the number of hands in the collection and the complexity of the handwriting.

As a general rule, we recommend adding 20 pages of Ground Truth every time you retrain a smaller model (as in the diary example above) and between 50 and 100 new pages with larger models. However, it may be good to experiment with adding different amounts of Ground Truth and see which is most effective for your model.

 <h3>How many times can you retrain a model?</h3> 

When it comes to how many you should retrain your model, there is no minimum or maximum. At the start of the retraining process, you should find that the CER of your model improves each time.

However, after a while, you will find that retraining your model stops having an effect on its CER. This happens because the model has already captured most of the information it can learn from the data, and additional data no longer leads to meaningful gains. At this point, you should stop retraining your model and take the most accurate version as the final version of your model.

 <img src="https://s3.eu-central-1.amazonaws.com/readcoop.cis.public-assets.prod/blog/training-models/Screenshot%202024-11-07%20121716.jpg" alt="Transkribus Help Center" style="max-width: 100%; height: auto;"> 

 <h2>Where can I find out more about retraining models?</h2> 

If you want more information about training text recognition models, Transkribus offers plenty of resources to help you deepen your understanding. For example, the <a href="https://help.transkribus.org/">Transkribus Help Center</a> has extensive guides on training different types of models, as well as on all the other tools Transkribus offers.

We also host regular free webinars on model training, transcription, layout analysis, and many other topics. Check out the <a href="https://www.transkribus.org/events">Events page</a> on the Transkribus website to register for these valuable sessions—they’re an excellent way to ask experts questions directly and learn best practices.

 <h2>Final thoughts</h2> 

Retraining a text recognition model in Transkribus might sound like a technical challenge, but once you’re familiar with the process, it becomes a useful way of continually improving the accuracy of your transcriptions. Each new version is a step closer to perfection, especially when you work with diverse, accurate Ground Truth data. Over time, you’ll be able to create a model that tackles even the most complex handwriting or unusual typefaces, helping to unlock your historical documents and preserve them in digital form.

How to Retrain a Model in Transkribus

How to retrain a model in Transkribus

Deciphering historical documents can be a daunting task, especially when faced with old Norwegian handwriting or typewritten texts from centuries past. Luckily, AI models have made significant progress in helping researchers, historians, and genealogists transcribe and understand these invaluable documents.

Whether you’re dealing with elegant script or typewritten letters, these four AI models offer powerful tools for transcribing old Norwegian texts. This post gives an overview of the specifics for each model and explains how they can enhance your transcription and digitisation projects.

 <img src="https://s3.eu-central-1.amazonaws.com/readcoop.cis.public-assets.prod/WP/uploads/2022/07/sunnhordland_partition_protocols-1.jpg" alt="Overview image of AI transcription models" style="width: 100%; height: auto;" /> 

<h3>1. Danish-Norwegian Handwriting (1874-1891)</h3>

Old handwritten documents present a unique challenge, particularly those from the late 19th century. The Danish-Norwegian Handwriting (1874-1891) model is specifically tailored to transcribe the elegant yet sometimes difficult-to-read handwriting from this period. Whether you’re working on family records, church registers, or official correspondence, this model will assist in transforming handwritten documents into digital text with ease.

Key Features:
<ul>
<li>Covers handwritten documents from 1874 to 1891.</li>
<li>Works on both Danish and Norwegian texts, expanding its usefulness for cross-border research.</li>
<li>Suited for transcription of official correspondence.</li>
</ul>

<a href="https://www.transkribus.org/model/danish-norwegian-handwriting-1874-1891-2">Try this model here</a>

 <img src="https://s3.eu-central-1.amazonaws.com/readcoop.cis.public-assets.prod/WP/uploads/2022/07/haugianerbrev_transcripts-1.jpg" alt="Image representing the Danish-Norwegian Handwriting model" style="width: 100%; height: auto;" /> 

<h3>2. Danish-Norwegian Typewritten Transcripts (1760-1842)</h3>

The Danish-Norwegian Typewritten Transcripts (1760-1842) was trained with 20th-century typewritten transcripts of handwritten letters from 1760-1842. This model is specifically designed to recognise both the typewritten characters as well as handwritten annotations and corrections.

Key Features:
<ul>
<li>Covers typewritten documents from the 20th century.</li>
<li>Optimised for the Norwegian and Danish languages, making it useful across both nations' historical archives.</li>
<li>Particularly effective for typewritten documents with handwritten corrections.</li>
</ul>

<a href="https://www.transkribus.org/model/danish-norwegian-typewritten-transcripts-1760-1842">Try this model here.</a>

 <img src="https://s3.eu-central-1.amazonaws.com/readcoop.cis.public-assets.prod/WP/uploads/2021/11/unnamed-1.png" alt="Image representing the Danish-Norwegian Typewritten Transcripts model" style="width: 100%; height: auto;" /> 

<h3>3. Norhand v0.7</h3>

The Norhand v0.7 model is a versatile tool, offering support for a wide range of handwritten Norwegian documents. Spanning two centuries, this model is highly adaptable and suited for various types of Norwegian handwriting, from everyday correspondence to formal records. It’s an excellent resource for anyone working with a diverse collection of historical Norwegian documents, especially when dealing with different handwriting styles.

Key Features:
<ul>
<li>Supports a broad range of Norwegian handwritten documents.</li>
<li>Particularly useful for research, allowing for the transcription of family records, parish registers, and personal letters.</li>
<li>Trained on documents from a 120-year timespan, accommodating the changing styles of handwriting over time.</li>
</ul>

<a href="https://www.transkribus.org/model/norhand_v0-7">Try this model here.</a>

 <h3>4. Norwegian Typewriting (1940-1945)</h3>

For those dealing with typewritten documents from the early to mid-20th century, the Norwegian Typewriting (1940-1945) model is specifically tuned to this era. This model was trained to transcribe Norwegian documents from the WWII era, making it perfect for historians studying the occupation of Norway, wartime correspondence, or even personal diaries that were typewritten during the war years.

Key Features:
<ul>
<li>Best suited for typewritten texts produced between 1940 and 1945.</li>
<li>Ideal for transcribing wartime documents, official letters, and other texts related to the WWII period in Norway.</li>
<li>Focuses on Norwegian language in a time of significant political and social upheaval.</li>
</ul>

<a href="https://www.transkribus.org/model/norwegian-typewriting-1940-1945">Try this model here.</a>

 <h3>How to use these AI models</h3>

Transkribus is an AI-powered platform that converts handwritten and printed historical documents into digital text.

Transkribus uses models such as these to create accurate transcriptions that are tailored to the language, style, and time period of your specific documents.

For more information about how to use AI models within Transkribus, please watch the video below or visit our <a href="https://help.transkribus.org/text-recognition">Help Center</a>.

Want to publish one of your models and share it with the community? You can find out how <a href="https://help.transkribus.org/managing-models">on this page</a>.

<div style="text-align: center;"><iframe width="560" height="315" src="https://www.youtube.com/embed/1y3ZABVnRmU" title="YouTube video player" frameborder="0" allow="accelerometer; autoplay; clipboard-write; encrypted-media; gyroscope; picture-in-picture" allowfullscreen></iframe></div>

4 AI Models for Reading Old Norwegian Handwriting and Typewritten Documents

<h1>Artificial intelligence in museums: three Transkribus case studies</h1>
Today, many industries are employing artificial intelligence (AI) to remain competitive and innovative. But what about museums? As institutions based around physical objects, museums may seem ill-suited to the digital age. However, they can derive significant benefits from integrating AI into their operations, enhancing access to museum collections and helping to manage extensive archives of written documents.
In this blog, we will explore how museums are leveraging AI and examine specific museum projects that have successfully implemented <a href="https://www.transkribus.org/">Transkribus’</a> AI tools.
<img src="https://s3.eu-central-1.amazonaws.com/readcoop.cis.public-assets.prod/blog/AI%20in%20museums/Image%2001.jpg" style="width:100%;height:auto;">
Technology plays an increasingly important role in the museum sector. © <a href="https://unsplash.com/de/@ishalyminov?utm_content=creditCopyText&amp;utm_medium=referral&amp;utm_source=unsplash">Igor Shalyminov</a>
<h2>From Machine Learning to Virtual Reality: the Evolving Role of Technology in Museums</h2>
Museums have traditionally relied on established methods for exhibiting and conserving their collections, which has, in part, led to them adopting AI technologies more slowly than other sectors. However, some pioneering museums have already begun to harness AI to enhance their services. For instance, some museums are now using AI to <a href="https://www.museumnext.com/article/artificial-intelligence-and-the-future-of-museums/">make exhibits more accessible</a> to visitors with hearing impairments and speakers of foreign languages. Other museums have created AI-powered virtual guides for their exhibits, while augmented reality is being increasingly <a href="https://www.youtube.com/watch?v=So_4UzvyPO4&amp;t=11s">incorporated into museum displays</a> to create more immersive and engaging experiences for visitors.
Many museums are also employing AI to digitise their extensive collections of written documents and further preserve these valuable materials. By making their collections digital, museums are not only safeguarding them for future generations but also making them more accessible to both researchers and the public. This increased accessibility ensures that the knowledge contained within these documents can be more widely shared, enriching our understanding of history and culture across society.
<img src="https://s3.eu-central-1.amazonaws.com/readcoop.cis.public-assets.prod/blog/AI%20in%20museums/Image%2002.jpg" style="width:100%;height:auto;">
Museums can use AI to digitise their extensive collections of handwritten documents. © <a href="https://unsplash.com/de/@annaauza?utm_content=creditCopyText&amp;utm_medium=referral&amp;utm_source=unsplash">Anna Auza</a>
<h2>Transkribus as a Digitisation Tool for Museums</h2>
One of the most advanced AI tools in this field is <a href="https://www.transkribus.org/">Transkribus</a>, which offers two primary benefits for museums: conservation and accessibility. Museums can upload scans of their documents and store them in a private, user-friendly Transkribus collection, where they can then be automatically transcribed, tagged, and searched using AI tools. Transkribus also provides a feature called <a href="https://www.transkribus.org/sites">Transkribus Sites</a>, an intelligible interface that allows museums to share their transcriptions with the research community and the general public.
In addition, the portable <a href="https://www.transkribus.org/scanning">Transkribus ScanTent</a> enables users to efficiently scan large quantities of documents using only a smartphone. This technology simplifies the conservation of archives and helps museums conduct the research necessary for new exhibits. In short, Transkribus facilitates the conservation and management of archives and improves their accessibility for museologists and visitors alike.
<img src="https://s3.eu-central-1.amazonaws.com/readcoop.cis.public-assets.prod/blog/AI%20in%20museums/Image%2003.jpg" style="width:100%;height:auto;">
The ScanTent enables documents to be scanned in a low-cost, efficient way. © Transkribus
<h2>The Future of Transkribus in Museums</h2>
In addition to the tools already available on the platform, the Transkribus team is actively developing innovative AI tools to facilitate the extraction of tangible, structured data from the unstructured information contained in documents. One such feature is named entity recognition, which will automatically identify key information within extensive sets of historical documents and organise it into a coherent structure. Manually performing this task would require significant time and resources, but AI makes it easily achievable.
Another tool that will soon be available on Transkribus is a specialised recognition model for tables. This will be particularly useful for museums that manage large collections of catalogues and other finding aids and need a quick, efficient way to digitally organise their content.
<img src="https://s3.eu-central-1.amazonaws.com/readcoop.cis.public-assets.prod/blog/AI%20in%20museums/Image%2004.JPG" style="width:100%;height:auto;">
Transkribus is currently developing several AI tools to assist with the digitisation of historical documents. © Transkribus
<h2>Spotlight on Success: Case Studies &amp; Examples</h2>
Let’s take a closer look at how Transkribus has been implemented into different museum operations:
<img src="https://s3.eu-central-1.amazonaws.com/readcoop.cis.public-assets.prod/blog/AI%20in%20museums/Image%2005.jpg" style="width:100%;height:auto;">
The Museum für Naturkunde is one of Berlin’s most prestigious museums. © Carola Radke, Museum für Naturkunde Berlin
<h3>Museum für Naturkunde</h3>
Berlin’s <a href="https://www.museumfuernaturkunde.berlin/en">Museum für Naturkunde</a> embarked on an ambitious <a href="https://www.museumfuernaturkunde.berlin/en/science/what-were-doing-collection-discovery-and-development-project">project</a> to digitise and extract data from over 250,000 insect specimen identification tags. By leveraging our named entity recognition model, the museum created a structured dataset from this enormous amount of information. This project is a major step toward enabling Collection Discovery and Development — by making the museum's biodiversity archives available to researchers and the public through its data portal.
<img src="https://s3.eu-central-1.amazonaws.com/readcoop.cis.public-assets.prod/blog/AI%20in%20museums/Image%2006.jpg" style="width:100%;height:auto;">
By using Transkribus, the Musée des Hospitalières de l’Hôtel-Dieu de Montréal can prepare more thoroughly for future exhibitions. © Gene.arboit via <a href="https://commons.wikimedia.org/wiki/File:Musee_des_Hospitalieres_de_l%27Hotel-Dieu_de_Montreal_2006-08-26.JPG">Wikimedia Commons</a>
<h3>Musée des Hospitalières</h3>
The <a href="https://museedeshospitalieres.qc.ca/">Musée des Hospitalières de l’Hôtel-Dieu de Montréal</a> has chosen Transkribus' AI tools to prepare for the renewal of its permanent exhibition, as well as for the creation of temporary ones. Using the keyword search function, the museum is systematically inventorying considerable masses of historical documents. This AI-driven approach is helping the museum share the heritage it preserves as exhaustively as possible.
<img src="https://s3.eu-central-1.amazonaws.com/readcoop.cis.public-assets.prod/blog/AI%20in%20museums/Image%2007.jpg" style="width:100%;height:auto;">
Technology is at the forefront of the National Museums of World Culture. © The World Culture Museums/Beatrice Törnros
<h3>National Museums of World Culture</h3>
In Sweden, the <a href="https://www.varldskulturmuseerna.se/en/">National Museums of World Culture</a> manages an astounding collection that includes over 500,000 objects, a million images, and hundreds of meters of archival material. To make these unique documents more accessible, the museums are using Transkribus Sites to publish large collections online for everyone to enjoy. This initiative allows both researchers and the general public to explore these invaluable archives from wherever they are in the world.
<h2>Find out more about Transkribus</h2>
Artificial intelligence is unlocking new possibilities for museums, helping them preserve their collections and engage with the public in innovative ways.
If you're interested in implementing Transkribus at your institution, then we are ready to help. You can try our <a href="http://Transkribus.ai">Transkribus.ai</a> demo version for free, or if you already have a project, then simply arrange a <a href="https://www.transkribus.org/contact/consultation">consultation with our team</a> to find out more about how Transkribus can help prepare your museum for the digital future.

This blog article was written by Frédéric Dagenais.

Artificial intelligence in museums: three Transkribus case studies

Comment lire de l’écriture française à l’aide de l’IA 

How to read French handwriting with AI

Explore Transkribus with Our Free Webinars!

<h1>New and Improved Field Models Now Available on Transkribus</h1> Field Models - our powerful, trainable models for structural layout recognition - are now live!After a successful beta phase, including continuous testing and valuable feedback from our community (who have already trained 500+ Field Models), we've fine-tuned this game-changing feature to be as efficient as possible. Whether you're working with newspapers, birth registers, or any other complex document, Field Models are here to improve your workflow and make it smarter.Today we're pleased to announce that the new and improved <a href="https://www.transkribus.org/en/field-models" target="_blank">Field Models</a> are now available in the Transkribus app as part of the Scholar, Team, and Organisation subscription plans.In this blog post, we'll give you a brief overview of the purpose and functionality of Field Models to help you decide if they're right for your project. We’ll also introduce you to our first two public Field Models, which are ideal for getting started with this new and exciting technology. <h2>The Importance of Layout Analysis</h2> Creating an automatic transcription in Transkribus involves two main steps. First, the platform carries out a layout analysis to determine where the text is located on the page. Once the text is identified, the platform then deciphers the characters and words, producing a transcription. This second step is referred to as text recognition.If your documents have a straightforward layout, such as a handwritten diary with straight lines of text, Transkribus can easily locate the text. However, documents with more complex layouts, such as newspapers with their multiple columns, varied text regions, headings, and embedded images, present a challenge for accurate layout extraction. Field Models help Transkribus identify and categorise these layout elements correctly, ensuring that the document's structure is preserved and understood in a semantically accurate manner.<img src="https://s3.eu-central-1.amazonaws.com/readcoop.cis.public-assets.prod/blog/field-models-launch/Grundbuchbl%C3%A4tter.jpg" alt="" style="max-width: 100%; height: 539.594px; width: 1021.07px;"> With a Field Model, you can train Transkribus to recognise the complex layout of your documents. © Die Tiroler Grundbuchblätter der k. u. k. Armee via Transkribus <h2>How Field Models Improve Layout Analysis</h2> Where layouts get complex is where Field Models become valuable. Field Models allow you to train a custom AI model to recognise complex layouts in your documents. Just as a text recognition model learns to recognise different handwriting styles, a Field Model learns to identify different layout elements, such as headlines, columns, or marginalia.You can also use Transkribus to detect specific elements within your documents, such as page numbers or headings, and add these to an existing layout. This is particularly useful if you've already processed a document but now want to label and focus on particular layout components. During text recognition, you can choose the structural regions identified by the Field Model and restrict the text recognition to just those areas.Additionally, you can train the model to recognise only certain parts of your document. For instance, if you need to extract only the heading and date, you can train a Field Model to identify these elements exclusively. When you run the text recognition, Transkribus will transcribe just these selected fields, eliminating the need to transcribe the entire page and then manually extract the desired parts. This targeted approach significantly streamlines your workflow, saving time and effort. <h2>An Example of How Field Models Work</h2> Let’s take a look at how this works in practice. In this example, you are wanting to digitise and transcribe a collection of newspaper index cards. Each card has several layout elements:<ul><li>a name</li><li>a shelfmark</li><li>the name of the newspaper</li><li>a detailed description</li><li>a reference</li><li>an “In Basement” label.</li></ul><span style="font-size: 11pt; font-family: Arial, sans-serif; color: rgb(0, 0, 0); background-color: transparent; font-variant-numeric: normal; font-variant-east-asian: normal; font-variant-alternates: normal; font-variant-position: normal; vertical-align: baseline; white-space-collapse: preserve;"><img src="https://lh7-rt.googleusercontent.com/docsz/AD_4nXfWCgARNASCXpm17mzPxBUM2y-OgwRpF7huIGnGjdp0Pm12uT4K6nFGzHXBzdAwt-rB5LGXpt7u0nyTfTzw2IyiHbCuCt2sr9haG9sI22k2428giMYKuEmvkXLYKaL3FTMrCSYlJZebqIJGnS8_CMA0xg?key=Cl5uAKQPBkN3qFKtTATZ2w" width="602" height="355" style="margin-left:0px;margin-top:0px;">The original index card. © Peabody Institute LibraryYou want to automatically transcribe everything on the index card except the “In Basement” label.If you run the text recognition using Transkribus’ default layout analysis settings and without a Field Model, you will be left with a transcription that looks like this:<span style="font-size: 11pt; font-family: Arial, sans-serif; color: rgb(0, 0, 0); background-color: transparent; font-variant-numeric: normal; font-variant-east-asian: normal; font-variant-alternates: normal; font-variant-position: normal; vertical-align: baseline; white-space-collapse: preserve;"><img src="https://lh7-rt.googleusercontent.com/docsz/AD_4nXe7xph8GxDxI-VQSE_XeFSswTG329E75UyS1YOugjE0IirMtbcZULIKA63A-m4IxE7gO0SYoMhfuBGn_fgFcqLhxvJURI9piJ0Rr_OJpOwLoUgmxSCK5u4hmNBL8TK3VOWyHt1qSp09PSBEN-EzfpY_LdA?key=Cl5uAKQPBkN3qFKtTATZ2w" width="602" height="231" style="margin-left:0px;margin-top:0px;"> Transcribed index card without a Field Model being applied first. © Peabody Institute LibraryWhen using the standard layout analysis in Transkribus, the text on an index card is transcribed as a continuous block without any structural information, treating all elements as part of the same undifferentiated text stream. This provides a basic transcription but lacks the detailed organisation necessary to accurately reflect the document’s structure. <span style="font-size: 11pt; font-family: Arial, sans-serif; color: rgb(0, 0, 0); background-color: transparent; font-variant-numeric: normal; font-variant-east-asian: normal; font-variant-alternates: normal; font-variant-position: normal; vertical-align: baseline; white-space-collapse: preserve;"><img src="https://lh7-rt.googleusercontent.com/docsz/AD_4nXftBuOOuOW2eWpjE7XEjbSBj_1BHLYjBI9ivXkiXd9CgryV4AzKP97rEi5jGvXqMSvM-Nhp3tFEfSgBNpf2-RtqUoFdaThQlIXMGdtJhf7sc-MtPwhNSjP6OYudILE00qvDR_Fr8Ed6h-5h0a9qMb_90D0?key=Cl5uAKQPBkN3qFKtTATZ2w" width="602" height="229" style="margin-left: 0px; margin-top: 0px; width: 617.055px; height: 234.718px;"> Index card after a Field Model has been applied. © Peabody Institute LibraryIn contrast, by applying a Field Model to the index card, Transkribus can recognise and preserve these distinct elements, such as the name, shelfmark, newspaper title, detailed description, and reference.Note that as the Field Model was trained to not recognise the “In Basement” label, it has simply ignored this layout element.<span style="font-size: 11pt; font-family: Arial, sans-serif; color: rgb(0, 0, 0); background-color: transparent; font-variant-numeric: normal; font-variant-east-asian: normal; font-variant-alternates: normal; font-variant-position: normal; vertical-align: baseline; white-space-collapse: preserve;"><img src="https://lh7-rt.googleusercontent.com/docsz/AD_4nXdE9Z-nM85H5ZbplWsKjP_B2YtgrJOzFE1FYuIlFAy9UviOkGDQXvy27ruzkgJ1Ytb5UH7vYNjqK0JmDt5zDEOLXLJPCfZcHKrHICcLvYv5-C3zzBIwsmKMwrvV-0CONS9RU0zbF8gYIEHtzSyJNB0XfA?key=Cl5uAKQPBkN3qFKtTATZ2w" width="602" height="227" style="margin-left: 0px; margin-top: 0px; width: 619.055px; height: 237.836px;"> Finished index card after layout analysis with a Field Model and text recognition. © Peabody Institute LibraryAs you can see above, this results in a transcription that not only includes the text but also maintains the layout’s structure. Unlike the previous example, no post-editing is needed to get this transcription up to scratch.The output is also enriched with structural metadata, making it much easier to use the information in a structured format, such as a database, where each element is clearly defined and categorised. This allows for more advanced data handling and analysis, ensuring that each part of the document is accurately represented and easily accessible. <h2>Which Other Kinds of Documents Are Field Models Useful For?</h2> In addition to the index cards shown above, there are many other document types that could benefit from the enhanced layout analysis that Field Models offer.<h3>Newspapers</h3>Newspapers are not only crucial for historical research, but, due to their complex but consistent layout, they are also excellent for use with Field Models. With accurate training data, a Field Model can learn to identify headlines, sub-headings, columns, images, and many other layout elements. Applying this layout analysis to large collections of newspapers can save a lot of time during the text recognition phase.<span style="font-size: 11pt; font-family: Arial, sans-serif; color: rgb(0, 0, 0); background-color: transparent; font-variant-numeric: normal; font-variant-east-asian: normal; font-variant-alternates: normal; font-variant-position: normal; vertical-align: baseline; white-space-collapse: preserve;"><img src="https://lh7-rt.googleusercontent.com/docsz/AD_4nXceTppC5EvFx1XAycw4erGdUgR_2PGFomqPJN6webgFw6i5Um7Lw07ecDtzIGhgHDFg21UHRCfh2jZZ5MB_m_maDgAAW_SjQZ6S2nNuXGrlRzMfgyU65TctiAH2yK123rmYMby5xQFeWc9TxDUR_1oQWvg?key=Cl5uAKQPBkN3qFKtTATZ2w" width="602" height="396" style="margin-left:0px;margin-top:0px;"> Newspapers are one of the document types most used with Transkribus. © KB, National Library of the Netherlands via Europeana <h3>Legal records and forms</h3>Legal records and forms contain consistent elements such as case numbers, names, locations and legal issues. Transkribus Field Models can automatically identify and extract these elements, speeding up the digitisation process. This helps reduce manual processing and increases the efficiency of research and document management.<h3>Scrapbooks</h3>Scrapbooks are, by nature, a collection of many different elements. From pictures and notes to transport tickets and newspaper clippings, they are limitless in the types of layout elements they can contain. With a Field Model, you can show Transkribus what all these elements look like, and train the AI to understand the layout of your individual scrapbook.<span style="font-size: 11pt; font-family: Arial, sans-serif; color: rgb(0, 0, 0); background-color: transparent; font-variant-numeric: normal; font-variant-east-asian: normal; font-variant-alternates: normal; font-variant-position: normal; vertical-align: baseline; white-space-collapse: preserve;"><img src="https://lh7-rt.googleusercontent.com/docsz/AD_4nXfWdsLdGdwKoJDOyjibdwPyztAtrT6j1TBvDvGyQUe8YVKVXFxHzfQS1ttxPqdKug_Feii1XRHKNLYnCuvcAlRDLaU2VYf-zUsUjUFTqLYsl0Hv7wZmszZhBmYaAdDwykcgQckD91gXjovLQVbVnMbAMw?key=Cl5uAKQPBkN3qFKtTATZ2w" width="602" height="485" style="margin-left:0px;margin-top:0px;"> Scrapbooks contain an unlimited number of layout elements. © Irish Traditional Music Archive via Europeana<h3>Sheet Music</h3>Applying Field Models to sheet music is currently the focus of several Digital Humanities research groups, who have successfully used Transkribus to separate the text and the notation elements. For more information, visit this article from <a href="https://www.br-klassik.de/aktuell/transkribus-kuenstliche-intelligenz-gregorianik-100.html">BR Klassik</a>. <h2>Are there Public Field Models?</h2> Yes, there are! We’re excited to share our first public Field Models, which provide ready-to-use tools to make your document processing easier.<h3>The Baroness of Blocks</h3> The first one, called "The Baroness of Blocks", is designed to help you identify printed text blocks as single regions instead of separate paragraphs. It's particularly useful for documents with clear column layouts, making it a great starting point for creating accurate ground truth data.<span style="font-size: 11pt; font-family: Arial, sans-serif; color: rgb(0, 0, 0); background-color: transparent; font-variant-numeric: normal; font-variant-east-asian: normal; font-variant-alternates: normal; font-variant-position: normal; vertical-align: baseline; white-space-collapse: preserve;"> <img src="https://lh7-rt.googleusercontent.com/docsz/AD_4nXeZX6skC6Md9B0kfhiuhcrIFtaC01e7EvM4yfJswr7_RLpvZedwwAsqR3Nf4yX-bVCAfcjhdw1hR1G-oTP5kKGGm_Dt5z72--egRBNqZYUYCrPkujHrnJwDh1m31irDq0AjnRqJRPw9bMgD3dm4bgZUkj8?key=Cl5uAKQPBkN3qFKtTATZ2w" width="602" height="344" style="margin-left: 0px; margin-top: 0px; width: 604.055px; height: 345.171px;"> The Baroness of Blocks © Transkribus<h3>The Marginalia Monarch</h3> Our next model, "The Marginalia Monarch", focuses on recognising marginalia - those handwritten or additional notes often found in the margins of documents. While it may need some adjustment for different types of marginalia, this model is a good place to start when it comes to extracting these important annotations.<span style="font-size: 11pt; font-family: Arial, sans-serif; color: rgb(0, 0, 0); background-color: transparent; font-variant-numeric: normal; font-variant-east-asian: normal; font-variant-alternates: normal; font-variant-position: normal; vertical-align: baseline; white-space-collapse: preserve;"><img src="https://lh7-rt.googleusercontent.com/docsz/AD_4nXc9lJhUc8QHPtToqrPaWNHH1xGFkOr7aQAPORutCnHsUJF_51Y3DS9Tb8tvLU-YatIhBa6Wvv--aVRuZx4WNbK2DbQWfZPr-W3UFXrMETZhv318nNgd3URWlhxNZYtQUC-xq-JHiSCtFlA9B5Rcqa_upw?key=Cl5uAKQPBkN3qFKtTATZ2w" width="602" height="344" style="margin-left:0px;margin-top:0px;"> Marginalia Monarch © Transkribus<h2>How do I train my own Field Model?</h2> Training a Field Model is a fairly straightforward process that doesn’t take too long to learn.In principle, it is similar to training a text recognition model. You first provide Ground Truth training data; in this case, documents in which you have already tagged the different layout elements. Transkribus uses these documents to train the model, which can then be applied to the rest of your collection.On our <a href="https://help.transkribus.org/field-models">Help Center</a>, you can find detailed instructions on how to train a Field Model for your documents, from preparing the training data to setting up the model.In addition, on September 3, we are holding a Field Models webinar, which will give you an overview of how to use Field Models and how to tailor them to your specific documents. There will also be time at the end to answer any questions you might have about the new technology.The webinar is free to attend. To sign up, please visit our <a href="https://www.transkribus.org/event/field-models-september-2024">Events page</a>.

New and Improved Field Models Now Available on Transkribus

Unlocking 5 Important Letters From History

<h1></h1>
The Netherlands is known not only for its rich history but also for the meticulous way it records that history. With state archives in every province, not to mention the famous National Archives of the Netherlands in Amsterdam, accessing Dutch documents such as birth registers and court records is relatively easy. However, reading the handwriting in these documents is often a little trickier. The combination of older styles of handwriting, outdated terms, and phrases, plus the fact that many documents may be partially damaged or illegible makes manual transcription a time-consuming task.
<img src="https://s3.eu-central-1.amazonaws.com/readcoop.cis.public-assets.prod/blog/5-models-dutch/Murakami_Gensui_dutch_writing.jpg" style="width: 100%; height: auto;">
Transkribus makes it much easier to read handwritten manuscripts in Dutch. © Murakami Gensui, Public domain, via Wikimedia Commons
Nowadays, though, AI can help. AI text recognition platforms such as Transkribus make it possible to automatically transcribe handwritten documents at the click of a button. You simply upload an image of the document, select an AI model on the platform, and receive an automatic transcription, which can then be edited and exported as required. Transkribus even lets you add metadata to your transcriptions, making it easier to search for named entities such as people or places.
If you want to transcribe your handwritten documents with Transkribus, here are five models that are ideal for Dutch handwriting. <h3 style="color: rgb(34, 34, 34); letter-spacing: 0.4px;">1) The Dutchess — an all-purpose model from the National Archives and the Amsterdam City Archives</h3> 
<img src="https://s3.eu-central-1.amazonaws.com/readcoop.cis.public-assets.prod/blog/5-models-dutch/dutchess.png" style="width: 100%; height: auto;">
The Dutchess is ideal for any kind of Dutch handwriting. © Transkribus
If you are using our <a href="https://www.transkribus.org/plans">free Individual plan</a>, then this is your go-to model for all things Dutch. Based on existing models from the National Archives and the <a href="https://archief.amsterdam/">Amsterdam City Archives</a>, among other institutions, the Dutchess is trained on a wealth of material from the 17th to the 19th century and is ideal for use with notarial writing, chronicles, and other such documents. You can find out more about the Dutchess <a href="https://app.transkribus.org/training/text/50786">here</a>. 
<h3>2) The Dutch Demeter — our premium Super Model for transcribing Dutch handwriting</h3><h3> </h3>
<img src="https://s3.eu-central-1.amazonaws.com/readcoop.cis.public-assets.prod/blog/5-models-dutch/demeter.jpg" style="width: 100%; height: auto;">
Our premium Dutch Demeter Super Model produces highly accurate transcriptions at lightning speed. © Transkribus
However, if you have upgraded to our <a href="https://www.transkribus.org/plans">Scholar plan</a>, then you have access to our powerful Dutch Demeter Super Model. Using advanced transformer technology, the Dutch Demeter offers unparalleled accuracy, producing complete transcriptions with little post-editing required. It was trained on a corpus of over 18 million words from materials provided by a range of institutions in the Netherlands and has a CER of 4.9%. You can find out more about the Dutch Demeter <a href="https://app.transkribus.org/training/text/58997">here</a>. 
<h3>3) Dutch 17th century — a specialist model for documents from the 1600s</h3><h3> </h3>
<img src="https://s3.eu-central-1.amazonaws.com/readcoop.cis.public-assets.prod/blog/5-models-dutch/Screenshot%202024-07-11%20093017.png" style="width: 100%; height: auto;">
17th-century documents such as these can be transcribed instantly. © Stadsarchief Gent via <a href="https://chronicling-novelty.com/kronieken-lezen/">Chronicling Novelty</a>
Created by a team from the Amsterdam City Archives and the <a href="https://chronicling-novelty.com/">Chronicling Novelty</a> research project, this model was trained exclusively on handwritten notarial documents from the 17th century. It includes materials from both Belgium and the Netherlands — over 1.8 million words in total. You can find out more about the Dutch 17th century model <a href="https://app.transkribus.org/training/text/38871">here</a>. 
<h3>4) Medieval Scripts - a large model for medieval texts and manuscripts</h3><h3> </h3>
<img src="https://s3.eu-central-1.amazonaws.com/readcoop.cis.public-assets.prod/blog/5-models-dutch/image%20%281%29.png" style="width: 100%; height: auto;">
Transkribus is also capable of reading medieval manuscripts such as this devotional book. © Universiteitsbibliotheek UGent, Public domain, via Wikimedia Commons
If you work with medieval documents in Dutch, then this is the model for you. Trained on over 7 million words, it is capable of reading not just handwriting in Dutch, but also in Latin, French, and German, and can transcribe documents from the entire medieval period. You can find out more about the Medieval Scripts model <a href="https://app.transkribus.org/training/text/42143">here</a>. 
<h3>5) War Letters - a dedicated model for 20th-century handwritten letters</h3><h3> </h3>
<img src="https://s3.eu-central-1.amazonaws.com/readcoop.cis.public-assets.prod/blog/5-models-dutch/Screenshot%202024-07-11%20094809.png" style="width: 100%; height: auto;">
The NIOD is one of many Dutch institutions that transcribe their documents with Transkribus. © NIOD Institute for War, Holocaust and Genocide Studies
Created by the <a href="https://www.niod.nl/en">NIOD Institute for War, Holocaust, and Genocide Studies</a> in Amsterdam, this unique model was trained on a collection of letters written between 1935 and 1950 by a wide range of individuals including children, soldiers, and Jewish people in hiding. It is ideal for transcribing handwritten letters in Dutch from this period. You can find out more about the War Letters model <a href="https://app.transkribus.org/training/text/53102">here</a>. 
<h2>How do I use an AI model with Transkribus?</h2>
Transkribus’ transcriptions are based on AI models. Each model has been trained to read a specific type of handwritten or printed text in a certain language, and often a certain time period or genre too.
To transcribe a document with Transkribus, you first need to upload a scan of the document and then you choose a model. There are over 200 public models available, which are all completely free to use. Transkribus will take the information stored in that model and apply it to your document, creating an instant transcription.
Sign up now for your free Transkribus account at <a href="https://account.readcoop.eu/auth/realms/readcoop/login-actions/registration?client_id=transkribus-webui-app&amp;tab_id=M0eCn6MM7-A">app.transkribus.org</a> or take a look at our First Steps video below to learn more.
If you are an archive or library and want to join the growing list of Dutch institutions using Transkribus, then <a href="https://www.transkribus.org/contact">set up a consultation</a> with us and find out how Transkribus could benefit your digitisation projects.
<a href="https://youtu.be/lCdL_kKBMn8" target="_blank">https://youtu.be/lCdL_kKBMn8</a>

5 AI models for transcribing Dutch handwriting

5 AI Models for Transcribing Letters and Documents in Dutch

 Introducing Table Models – Trainable Layout AI in Transkribus

Think of an archive and you might think of old, dusty basements full of disorganised boxes of papers. But in the technological age, archives are taking on a new role. No longer are they purely physical collections of papers, manuscripts, or physical media, such as photographs or artwork. Instead, archives are becoming fully searchable digital collections, making it easier for researchers and the general public to access historical records and items from anywhere, at any time.



However, digitising an archive in this way is a challenge. A large national archive can hold millions of pages of text, which all have to be transcribed in order to produce digital versions. Previously, this was done manually — each individual word in the document was typed into a computer program by a human being. As you can imagine, transcribing in this way is a very time-consuming process and it would take literally decades to transcribe the millions of archival materials in a large archive.



That’s where Transkribus comes in. Transkribus is an AI-powered platform that uses machine learning to learn how to read a specific type of handwriting, no matter which language the text is in or when it was written. Once the platform can sufficiently recognise the handwriting, it can then transcribe thousands of pages at the click of a button.



In this post, we would like to give you an overview of the different kinds of archival materials which can be transcribed using Transkribus. You will learn how to process materials using the platform and there are also examples of archives which have successfully digitised their collections with Transkribus and made them fit for the future.



<figure class="wp-block-image"><img decoding="async" src="https://upload.wikimedia.org/wikipedia/commons/thumb/8/87/Archives_du_chapitre_de_Sion_01.jpg/640px-Archives_du_chapitre_de_Sion_01.jpg" alt="" style="width: 475.92px; height: 634.312px;"></figure>



Archives house many different types of written material. © <a target="_blank" rel="noreferrer noopener" href="https://commons.wikimedia.org/wiki/User:Tatakdh">Tatakdh</a> / Wikimedia Commons / CC BY-SA 4.0



<h2 class="wp-block-heading" id="h-what-are-archival-materials">What are archival materials?</h2>



There are many different types of archival materials but they all have one thing in common: they are all primary sources. This means they were written at the time a particular historical event took place, usually by someone who was there. Because of this, archival materials are particularly useful for historical research and most items in an archive can be used as scholarly sources. Some of the key types of archival materials include:



Manuscripts: These are any kind of material that is written by hand (“manu scripti” is Latin for “handwritten”). The oldest manuscripts still surviving today are thousands of years old and extremely fragile. Handling archival material such as old manuscripts is a delicate task and so creating digital versions of them makes it possible for everyone to access the contents without damaging the manuscript.



Archival records: These are documents written by individuals or organisations to keep a written account of events such as births, deaths, meetings, finances, and many others. According to the <a target="_blank" rel="noreferrer noopener" href="https://dictionary.archivists.org/entry/archival-record.html">SAA Glossary</a>, the term “archival record” usually denotes unpublished documents, rather than published artefacts such as books.



Letters, diaries, and personal papers: These very important types of material give an insight into the lives of the people who wrote them and the world they were living in at the time. They are particularly useful when studying a single individual or small group.



Organising archival materials is more than just writing a description and putting it on the right shelf. Each item is classified according to several “elements”, which are included in the archival description. These include intellectual elements, such as the name, date and type of material; physical occurrence elements such as the number of pages or location; and the media occurrence element, which covers the media type, colour, dimensions, piece count, and reproduction count. These archival materials elements make it easier for someone to find a certain material and immediately understand its context.



<figure class="wp-block-image size-full"><img decoding="async" src="https://s3.eu-central-1.amazonaws.com/readcoop.cis.public-assets.prod/WP/uploads/2021/04/scantent.jpg" alt="" class="wp-image-23119" srcset="https://s3.eu-central-1.amazonaws.com/readcoop.cis.public-assets.prod/WP/uploads/2021/04/scantent.jpg 801w, https://s3.eu-central-1.amazonaws.com/readcoop.cis.public-assets.prod/WP/uploads/2021/04/scantent-300x166.jpg 300w, https://s3.eu-central-1.amazonaws.com/readcoop.cis.public-assets.prod/WP/uploads/2021/04/scantent-768x425.jpg 768w, https://s3.eu-central-1.amazonaws.com/readcoop.cis.public-assets.prod/WP/uploads/2021/04/scantent-16x9.jpg 16w, https://s3.eu-central-1.amazonaws.com/readcoop.cis.public-assets.prod/WP/uploads/2021/04/scantent-600x332.jpg 600w" sizes="(max-width: 801px) 100vw, 801px" ></figure>



The ScanTent can be used to create high-quality scans of archival materials. © Transkribus



<h2 class="wp-block-heading" id="h-a-step-by-step-guide-to-digitising-archival-materials-with-transkribus">A step-by-step guide to digitising archival materials with Transkribus</h2>



Transkribus is an AI-powered platform which converts handwritten or printed text into digital transcriptions. It is used in many different archives around the world to transcribe materials in just a fraction of the time it would take a human being. This allows whole collections to be quickly digitised and made accessible to everyone.



Below is a short overview of how to transcribe materials with Transkribus. For more details, check out our <a target="_blank" rel="noreferrer noopener" href="https://help.transkribus.com/">Help Center</a>, which contains detailed information on each step of the process.



<h3 class="wp-block-heading" id="h-step-1-create-images-of-the-archival-materials">Step 1: Create images of the archival materials</h3>



To use Transkribus, you need high-quality images or scans of the documents you want to transcribe. These should be in JPEG, PNG or PDF format.



You can create images using anything from sophisticated scanning technology to just your regular smartphone. If you are using this second option, we recommend using a device such as the <a target="_blank" rel="noreferrer noopener" href="https://readcoop.eu/scantent/">ScanTent</a>, to ensure optimal lighting and image quality.



<h3 class="wp-block-heading" id="h-step-2-sign-up-to-transkribus">Step 2: Sign up to Transkribus</h3>



Before you can do any transcribing in Transkribus, you first need to create an account. Go to <a href="http://app.transkribus.eu" target="_blank" rel="noreferrer noopener">app.transkribus.eu</a> , click on “Sign up for free” and follow the instructions. Find out more about <a href="https://help.transkribus.com/first-steps-in-transkribus" target="_blank" rel="noreferrer noopener">registration and login</a> in our Help Center.



<h3 class="wp-block-heading" id="h-step-3-upload-your-images">Step 3: Upload your images</h3>



Images are stored in “collections” on Transkribus. You can organise collections however you like. For example, a collection could include all the scans from a certain book or physical collection.



Start your collection by clicking “Tools” and then “Create a collection”. Once the collection has been created, select it on your Workdesk and then click “Upload” from the left-hand toolbar. You can then select the documents you wish to upload.



Your images and data will be stored on the servers of READ-COOP SCE, which are all located in Innsbruck, Austria, in a GDPR-compliant manner, and may be processed according to our&nbsp;<a href="https://readcoop.eu/terms-and-conditions/" target="_blank" rel="noreferrer noopener">terms &amp; conditions</a>. If data cannot leave your infrastructure for privacy reasons, check our On-Prem solution (<a href="https://readcoop.eu/transkribus/on-prem/" target="_blank" rel="noreferrer noopener">https://readcoop.eu/transkribus/on-prem/</a>).



Find out more about&nbsp;<a target="_blank" href="https://help.transkribus.com/creating-a-collection?__hstc=175940285.f7e3f39fd2fcee05fb32b06556209a24.1676363691689.1681200919014.1681204556052.72&amp;__hssc=175940285.1.1681204556052&amp;__hsfp=179051687" rel="noreferrer noopener">creating collections</a>&nbsp;and&nbsp;<a target="_blank" href="https://help.transkribus.com/uploading-files-to-transkribus?__hstc=175940285.f7e3f39fd2fcee05fb32b06556209a24.1676363691689.1681200919014.1681204556052.72&amp;__hssc=175940285.1.1681204556052&amp;__hsfp=179051687" rel="noreferrer noopener">uploading documents</a>&nbsp;in our Help Center.



<h3 class="wp-block-heading" id="h-step-4-create-an-automatic-transcription">Step 4: Create an automatic transcription</h3>



Open your collection and select a particular image. Click “Start automatic transcription” to begin the transcription process.



You will then need to select a model. This is a piece of code that contains all the information Transkribus needs to transcribe the text accurately. There are over 100 public models available, covering different languages and time periods. Select the one that is most relevant to your material and then click “Start”.



Once the processing is complete, the automatic transcription will appear on the right side of the screen. Find out more about <a href="https://help.transkribus.com/automatically-transcribing-your-documents" target="_blank" rel="noreferrer noopener">automatically transcribing documents</a> in our Help Center.



<figure class="wp-block-image size-large"><img decoding="async" src="https://s3.eu-central-1.amazonaws.com/readcoop.cis.public-assets.prod/WP/uploads/2023/04/Screenshot-2023-03-31-105613-1024x498.jpg" alt="" class="wp-image-32210" srcset="https://s3.eu-central-1.amazonaws.com/readcoop.cis.public-assets.prod/WP/uploads/2023/04/Screenshot-2023-03-31-105613-1024x498.jpg 1024w, https://s3.eu-central-1.amazonaws.com/readcoop.cis.public-assets.prod/WP/uploads/2023/04/Screenshot-2023-03-31-105613-300x146.jpg 300w, https://s3.eu-central-1.amazonaws.com/readcoop.cis.public-assets.prod/WP/uploads/2023/04/Screenshot-2023-03-31-105613-768x374.jpg 768w, https://s3.eu-central-1.amazonaws.com/readcoop.cis.public-assets.prod/WP/uploads/2023/04/Screenshot-2023-03-31-105613-1536x747.jpg 1536w, https://s3.eu-central-1.amazonaws.com/readcoop.cis.public-assets.prod/WP/uploads/2023/04/Screenshot-2023-03-31-105613-18x9.jpg 18w, https://s3.eu-central-1.amazonaws.com/readcoop.cis.public-assets.prod/WP/uploads/2023/04/Screenshot-2023-03-31-105613-600x292.jpg 600w, https://s3.eu-central-1.amazonaws.com/readcoop.cis.public-assets.prod/WP/uploads/2023/04/Screenshot-2023-03-31-105613.jpg 1916w" sizes="(max-width: 1024px) 100vw, 1024px" ></figure>



Transkribus can create automatic transcriptions of any historical document. © Transkribus



<h2 class="wp-block-heading" id="h-how-to-improve-the-accuracy-and-efficiency-of-your-transcriptions">How to improve the accuracy and efficiency of your transcriptions</h2>



Public models like the ones described above are trained to transcribe a broad range of archival materials with reasonable accuracy. However, if you want to improve the accuracy or efficiency of your transcriptions, here are a couple of tips.



<h3 class="wp-block-heading" id="h-check-the-quality-of-your-scans">Check the quality of your scans</h3>



The better the quality of your scans, the better the automatic transcriptions will be. Therefore, it makes sense to check that all the text on your scans is clearly legible — as mentioned above, good lighting when taking the images helps a lot here. You should also make sure that there are no marks or other “noise” which might affect how Transkribus processes them.



Finally, check that all the scans were uploaded with the correct orientation (portrait or landscape), as this will also have an impact on the layout recognition. Rotate any incorrect scans back to their original orientation and re-upload them to Transkribus.



<h3 class="wp-block-heading" id="h-improve-layout-recognition">Improve layout recognition</h3>



Before the text can be recognised, Transkribus first needs to know where on the page the text is located. It does this using layout recognition. This analyses which parts of the document contain text and visibly marks each individual line. It is these lines of text which are then recognised by the platform.



However, that also means that if the layout recognition is incorrect, then the text will also be incorrectly transcribed. Therefore, any errors in the layout recognition must be corrected manually to ensure an accurate transcription. If you have a material based on a table — for example, a birth register — then you can also programme this separately into Transkribus, so that the platform recognises the rows and columns correctly.



Find out more about <a target="_blank" rel="noreferrer noopener" href="https://help.transkribus.com/manual-layout-recognition">manual layout recognition</a> and <a target="_blank" rel="noreferrer noopener" href="https://help.transkribus.com/tables">tables</a> in our Help Center.



<figure class="wp-block-image size-full"><img decoding="async" src="https://s3.eu-central-1.amazonaws.com/readcoop.cis.public-assets.prod/WP/uploads/2023/04/Screenshot-2023-03-31-110537.jpg" alt="" class="wp-image-32211" srcset="https://s3.eu-central-1.amazonaws.com/readcoop.cis.public-assets.prod/WP/uploads/2023/04/Screenshot-2023-03-31-110537.jpg 953w, https://s3.eu-central-1.amazonaws.com/readcoop.cis.public-assets.prod/WP/uploads/2023/04/Screenshot-2023-03-31-110537-300x189.jpg 300w, https://s3.eu-central-1.amazonaws.com/readcoop.cis.public-assets.prod/WP/uploads/2023/04/Screenshot-2023-03-31-110537-768x483.jpg 768w, https://s3.eu-central-1.amazonaws.com/readcoop.cis.public-assets.prod/WP/uploads/2023/04/Screenshot-2023-03-31-110537-18x12.jpg 18w, https://s3.eu-central-1.amazonaws.com/readcoop.cis.public-assets.prod/WP/uploads/2023/04/Screenshot-2023-03-31-110537-600x377.jpg 600w" sizes="(max-width: 953px) 100vw, 953px" ></figure>



Good layout recognition leads to good text recognition. © Transkribus



<h3 class="wp-block-heading" id="h-train-a-custom-model">Train a custom model</h3>



Transkribus doesn’t just provide public models, it allows you to train custom models, too. These are models that are tailor-made to your individual materials, resulting in more accurate transcriptions.



To train a custom model, you first need to manually transcribe around 5,000-15,000 words, depending on the complexity of the handwriting. Transkribus will then use this transcribed material to learn how to read the handwriting and create a model so that it can recognise similar texts in the future.



Many archives train custom models to recognise a specific type of collection, for example, council records from a particular time period or notarial deeds written by a select group of notaries. This allows the collections to be transcribed much more accurately, with less post-editing required afterwards. Find out more about <a target="_blank" rel="noreferrer noopener" href="https://help.transkribus.com/training-models">training models</a> in our Help Center.



<h2 class="wp-block-heading" id="h-3-archives-that-have-successfully-digitised-their-materials-with-transkribus">3 archives that have successfully digitised their materials with Transkribus</h2>



Archives around the world have used Transkribus to transcribe many different types of materials and create digital versions of them. For more information about some of our most successful archival projects, check out the blog articles below.



<figure class="wp-block-image"><img decoding="async" src="https://s3.eu-central-1.amazonaws.com/readcoop.cis.public-assets.prod/WP/uploads/2022/10/NAN.jpg" alt="" s></figure>



<h3 class="wp-block-heading" id="h-transcribing-3-million-scans-at-the-national-archives-of-the-netherlands">Transcribing 3 million scans at the National Archives of the Netherlands</h3>



The National Archives of the Netherlands is home to literally millions of pages of material, both printed and handwritten. As part of their digitisation strategy, the archives used Transkribus to transcribe 3 million pages of records and notarial deeds from the 17th-19th centuries. The custom model they trained is also now available to all as a public model. Find out more about the project <a target="_blank" rel="noreferrer noopener" href="https://readcoop.eu/success-stories/national-archives-of-the-netherlands/">here</a>.



<figure class="wp-block-image"> </figure>



<figure class="wp-block-image"><img decoding="async" src="https://s3.eu-central-1.amazonaws.com/readcoop.cis.public-assets.prod/WP/uploads/2022/11/zuerich.jpg" alt="" ></figure>



<h3 class="wp-block-heading" id="h-how-the-state-archives-of-zurich-published-50-000-pages-online">How the State Archives of Zurich published 50,000 pages online</h3>



The State Archives of Zurich is a mine of historical information about the Swiss city. However, physically accessing the archive’s premodern collection was a challenge, prompting the archive to create a digital version which could be easily accessed online. Thanks to an accurate model, the team was able to transcribe and publish 50,000 pages of council minutes in just three years. Find out more about the project <a target="_blank" rel="noreferrer noopener" href="https://readcoop.eu/success-stories/state-archives-of-zurich/">here</a>.



<h2 class="wp-block-heading" id="h-try-transkribus-for-yourself">Try Transkribus for yourself</h2>



Transkribus is an AI-powered platform that transcribes handwritten text at the push of a button.



You can try the full version of Transkribus at <a href="http://app.transkribus.eu" target="_blank" rel="noreferrer noopener">app.transkribus.eu</a> or try our demo version below.




<div id="t-ai-app" data-embed="1" data-lang="en">
 <script src="https://transkribus.ai/t-ai-inline.js"></script>
</div>

How to digitise archival materials with Transkribus

Creating books in the Middle Ages was a challenge. Before the introduction of the printing press, all books were painstakingly copied out by hand and those such as the Bible could take several years to complete.



One particularly famous type of manuscript from the medieval period is the illuminated manuscript. The words in these manuscripts were embellished with ornate borders, detailed illustrations, and gold-leaf work, turning them into beautiful pieces of medieval art. Better still, many of them have survived to the present day, making them a rich source of information for medievalists and other historians.



In this post, we are going to take a look at what exactly an illuminated manuscript is and find out how these ornate medieval papers are being brought into the digital age.



<h2 class="wp-block-heading" id="h-what-is-an-illuminated-manuscript">What is an illuminated manuscript?</h2>



Illuminated manuscripts were a type of manuscript that were produced mainly in Western Europe from the 6th century until the invention of the printing press in the 15th century. Like all manuscripts, they were written by hand on a page of parchment, usually a piece of animal skin known as &#8220;vellum&#8221;. Many illuminated manuscripts were then bound into an early type of book called a &#8220;codex&#8221; (&#8220;codices&#8221; in the plural). A codex consisted of numerous sheets of parchment sandwiched between two pieces of wood, just like a modern book has pages bound between the covers. Luckily for today&#8217;s medieval historians, the wood preserved the pages inside, keeping them in peak condition and making it possible to still read the manuscripts today.



But what sets illuminated manuscripts apart from other medieval documents are the ornate illustrations, or &#8220;illuminations&#8221;, that adorn each page. Illuminations ranged from decorative borders around the page to a full-page illustration depicting an event described in the text. The first letter on a manuscript page — the &#8220;illuminated capital&#8221; — was also often enlarged and colourfully decorated. Often, this decoration was then enhanced with gold or silver leaf, especially if the manuscript was of particular importance. These shiny materials were said to &#8220;illuminate&#8221; the text, giving the manuscripts their name.



<figure class="wp-block-image size-large"><img decoding="async" width="1024" height="666" src="https://s3.eu-central-1.amazonaws.com/readcoop.cis.public-assets.prod/WP/uploads/2023/03/medieval-manuscript-1400x900-1-1024x666.png" alt="" class="wp-image-32035" srcset="https://s3.eu-central-1.amazonaws.com/readcoop.cis.public-assets.prod/WP/uploads/2023/03/medieval-manuscript-1400x900-1-1024x666.png 1024w, https://s3.eu-central-1.amazonaws.com/readcoop.cis.public-assets.prod/WP/uploads/2023/03/medieval-manuscript-1400x900-1-300x195.png 300w, https://s3.eu-central-1.amazonaws.com/readcoop.cis.public-assets.prod/WP/uploads/2023/03/medieval-manuscript-1400x900-1-768x499.png 768w, https://s3.eu-central-1.amazonaws.com/readcoop.cis.public-assets.prod/WP/uploads/2023/03/medieval-manuscript-1400x900-1-18x12.png 18w, https://s3.eu-central-1.amazonaws.com/readcoop.cis.public-assets.prod/WP/uploads/2023/03/medieval-manuscript-1400x900-1-600x390.png 600w, https://s3.eu-central-1.amazonaws.com/readcoop.cis.public-assets.prod/WP/uploads/2023/03/medieval-manuscript-1400x900-1.png 1400w" sizes="(max-width: 1024px) 100vw, 1024px" /><figcaption>The Bird Psalter, London (1284) © The Fitzwilliam Museum, Cambridge</figcaption></figure>







<h2 class="wp-block-heading" id="h-how-were-illuminated-manuscripts-made">How were illuminated manuscripts made?</h2>



The first illuminated manuscripts were created in monasteries in Northern Europe, most notably in the British Isles and Ireland. Most monasteries at the time had not just a library but also a &#8220;scriptorium&#8221;, where manuscripts and books were created.



Some monks worked as scribes. They were responsible for preparing and cutting the vellum and adding the text to the page in a certain colour of ink. If more colours were required, the page would then be passed to another scribe working with that colour. Other scribes in the room would be responsible for proofreading the text and checking that everything had been copied correctly. Once the scribes were finished with the text, the page was handed over to another monk, known as the &#8220;illuminator&#8221;, who worked essentially as an artist. He would add the illustrations to the page, as well as any gold leaf decoration that was required.



By the late medieval period, monks were no longer the only people creating illuminated manuscripts. With the rise of universities and the middle class, books started to become more popular and book production was increasingly more profitable. Therefore, booksellers in cities started to hire their own scribes and illuminators to create books outside of the monastery context.



<figure class="wp-block-image"><img decoding="async" src="https://upload.wikimedia.org/wikipedia/commons/thumb/0/0e/Museo_dell%27Opera_del_Duomo_%28Florence%29_-_Illuminated_manuscript-3.jpg/640px-Museo_dell%27Opera_del_Duomo_%28Florence%29_-_Illuminated_manuscript-3.jpg" alt=""/></figure>



Musical notation was also illuminated. © <a target="_blank" rel="noreferrer noopener" href="https://commons.wikimedia.org/wiki/User:Yair-haklai">Yair Haklai</a> / Wikimedia Commons / CC BY-SA 4.0



<h2 class="wp-block-heading" id="h-what-kind-of-illuminated-manuscripts-are-there">What kind of illuminated manuscripts are there?</h2>



As you might expect from a monastery, the majority of illuminated manuscripts are religious in nature. This included Bibles, copies of certain gospels or religious texts, and most commonly, books of hours. A book of hours was a collection of prayers, texts, and psalms, organised into the different canonical hours of the day. They became enormously popular during the late medieval period when rich families would commission individual collections for their households. Due to the sheer numbers created during that period, many books of hours have survived until the present day.



But illuminated manuscripts were also created on secular topics, too. Visual topics, for example, herbs or astrology, were the ideal material for illuminated books, and many such manuscripts were produced. Illuminated texts were also created about the lives of saints or famous adventurers, particularly towards the end of the Middle Ages.



<figure class="wp-block-image"><img decoding="async" src="https://upload.wikimedia.org/wikipedia/commons/thumb/2/2e/Book_of_hours_by_the_Master_of_Zweder_van_Culemborg_-_KB_79_K_2_-_folios_112v_%28left%29_and_113r_%28right%29.jpg/640px-Book_of_hours_by_the_Master_of_Zweder_van_Culemborg_-_KB_79_K_2_-_folios_112v_%28left%29_and_113r_%28right%29.jpg" alt=""/></figure>



The book of hours by the Master(s) of Zweder van Culemborg (public domain)



<h2 class="wp-block-heading" id="h-why-are-the-margins-so-important-in-illuminated-manuscripts">Why are the margins so important in illuminated manuscripts?</h2>



If you search online for information about illuminated manuscripts, you will find that many people talk about the &#8220;marginalia&#8221;. For historians, it is often the details included in these margins that are particularly interesting.



Creating manuscripts was difficult work. Scribes worked long hours, usually in silence, and copying the same text all day long was tedious. Therefore, they would often leave <a target="_blank" rel="noreferrer noopener" href="https://www.theatlantic.com/entertainment/archive/2012/03/cheeky-complaints-monks-scribbled-in-the-margins-of-manuscripts/254868/">personal comments in the margins</a>, such as: &#8220;The ink is thin&#8221;, &#8220;I am very cold&#8221;, or more poetically, &#8220;As the harbour is welcome to the sailor, so is the last line to the scribe.&#8221;



Perhaps even more shocking are the small illustrations that appear in the margins of medieval illuminated manuscripts. These often included nudity or scenes of a sexual nature, which were very much at odds with the religious content on the page! Another common theme in marginalia was illustrations of animals, particularly of animals doing human activities such as baking bread or playing instruments. Scholars believe these were often drawn as a form of artistic parody.



One animal-themed manuscript illumination frequently found in marginalia is the depiction of a knight fighting a snail. These little illustrations can be found in various manuscripts throughout the 13th and 14th centuries, and there are <a target="_blank" rel="noreferrer noopener" href="https://blogs.bl.uk/digitisedmanuscripts/2013/09/knight-v-snail.html">several ideas about their meaning</a>. Some scholars believe the snails represent the Lombards, a group vilified in medieval times for their unchivalrous behaviour. Others say the illuminations symbolise the struggles of the poor against an oppressive aristocracy. Whatever the reason, it is details like these that make a medieval manuscript so fascinating to study.



<figure class="wp-block-image"><img decoding="async" src="https://blogs.bl.uk/.a/6a00d8341c464853ef019aff95ab29970c-500wi" alt=""/></figure>



Knight vs Snail from the Gorleston Psalter © British Library



<h2 class="wp-block-heading" id="h-which-are-the-most-famous-manuscripts">Which are the most famous manuscripts?</h2>



There are thousands of illuminated manuscripts still around today, which can be viewed in an institution such as a museum or library. However, some manuscripts hold a particularly important place in history. Here are three examples:



<h3 class="wp-block-heading" id="h-the-book-of-kells">The Book of Kells</h3>



This 9th-century manuscript is probably one of the most famous around the world. It contains the four gospels of the New Testament, adorned with many different illuminations, including Celtic knotwork typical of County Meath, where the book was stored for centuries.



Nowadays, the Book of Kells can be seen at <a target="_blank" rel="noreferrer noopener" href="https://www.tcd.ie/">Trinty College Dublin</a>, who are a member of <a target="_blank" rel="noreferrer noopener" href="https://readcoop.eu/about/">READ-COOP</a>. Only certain pages of the manuscript are on display at any one time, and these are rotated frequently. It can also be viewed online on the <a target="_blank" rel="noreferrer noopener" href="https://digitalcollections.tcd.ie/collections/ks65hc20t">Trinity College website</a>.



<figure class="wp-block-image"><img decoding="async" src="https://upload.wikimedia.org/wikipedia/commons/thumb/b/b6/KellsFol114rArrestOfChrist.jpg/640px-KellsFol114rArrestOfChrist.jpg" alt=""/></figure>



Christ&#8217;s arrest in the Book of Kells (Folio 114v) © Trinity College Dublin



<h3 class="wp-block-heading" id="h-the-lindisfarne-gospels">The Lindisfarne Gospels</h3>



Written in the 8th century, this renowned illuminated manuscript was produced at the island monastery of Lindisfarne, in the north of England. Like the Book of Kells, it is a copy of the four gospels and is a prime example of medieval manuscript art. Over 90 colours were produced for the book, with gold used in just a few places.



The Lindisfarne Gospels are housed at the <a target="_blank" rel="noreferrer noopener" href="https://www.bl.uk/">British Library</a> in London, who are also a <a target="_blank" rel="noreferrer noopener" href="https://readcoop.eu/about/">READ-COOP</a> member.



<img decoding="async" src="https://blogs.bl.uk/.a/6a00d8341c464853ef02af14aa7b9f200b-500wi">A page from the Lindisfarne Gospels. © British Library



<h3 class="wp-block-heading" id="h-westminster-abbey-bestiary">Westminster Abbey Bestiary</h3>



In the Middle Ages, bestiaries were one of the most popular types of illuminated manuscript. Officially an encyclopedia of the animal kingdom, they often included fantastical beasts as well as actual creatures, accompanied by an explanation of the animal&#8217;s Christian significance. The illuminations in these manuscripts were sometimes so famous that they were replicated in other types of medieval art, such as carvings and tapestries.



The Westminster Abbey Bestiary is a fine example of a bestiary. Believed to have been produced in York at the end of the 14th century, it contained 164 miniatures of animals with Latin descriptions alongside. The bestiary is preserved by the <a target="_blank" rel="noreferrer noopener" href="https://www.westminster-abbey.org/about-the-abbey/library-research">Library of Westminster Abbey</a> in London and can be viewed on request.



<figure class="wp-block-image size-large"><img decoding="async" width="1024" height="718" src="https://s3.eu-central-1.amazonaws.com/readcoop.cis.public-assets.prod/WP/uploads/2023/03/FabogbTWYAA5T46-1024x718.jpeg" alt="" class="wp-image-32036" srcset="https://s3.eu-central-1.amazonaws.com/readcoop.cis.public-assets.prod/WP/uploads/2023/03/FabogbTWYAA5T46-1024x718.jpeg 1024w, https://s3.eu-central-1.amazonaws.com/readcoop.cis.public-assets.prod/WP/uploads/2023/03/FabogbTWYAA5T46-300x210.jpeg 300w, https://s3.eu-central-1.amazonaws.com/readcoop.cis.public-assets.prod/WP/uploads/2023/03/FabogbTWYAA5T46-768x538.jpeg 768w, https://s3.eu-central-1.amazonaws.com/readcoop.cis.public-assets.prod/WP/uploads/2023/03/FabogbTWYAA5T46-1536x1077.jpeg 1536w, https://s3.eu-central-1.amazonaws.com/readcoop.cis.public-assets.prod/WP/uploads/2023/03/FabogbTWYAA5T46-2048x1436.jpeg 2048w, https://s3.eu-central-1.amazonaws.com/readcoop.cis.public-assets.prod/WP/uploads/2023/03/FabogbTWYAA5T46-18x12.jpeg 18w, https://s3.eu-central-1.amazonaws.com/readcoop.cis.public-assets.prod/WP/uploads/2023/03/FabogbTWYAA5T46-600x421.jpeg 600w" sizes="(max-width: 1024px) 100vw, 1024px" /><figcaption>A fantastical creature from the Westminster Abbey Bestiary. © Westminster Abbey</figcaption></figure>







<h2 class="wp-block-heading" id="h-how-do-you-digitise-an-illuminated-manuscript">How do you digitise an illuminated manuscript?</h2>



Thanks to handwriting recognition technology like <a target="_blank" rel="noreferrer noopener" href="https://readcoop.eu/transkribus/?sc=Transkribus">Transkribus</a>, it is becoming easier and easier to create digital versions of historical manuscripts. Many museums and libraries have already scanned their collections of illuminated manuscripts and made them publically available online, for example, the <a target="_blank" rel="noreferrer noopener" href="https://www.bl.uk/catalogues/illuminatedmanuscripts/welcome.htm">Catalogue of Illuminated Manuscripts</a> at the British Library. These scans can then be uploaded to Transkribus, to create an automatic transcription of the text.



Illuminated manuscripts can be challenging to text recognition platforms, which sometimes have difficulty distinguishing between text and images. However, with Transkribus' <a href="https://www.transkribus.org/field-models">Field Models</a>, you can now teach the platform to distinguish between text and images, resulting in more accurate transcriptions that better reflect the layout of the page.



<h2>Are there AI models for medieval manuscripts?</h2>Transkribus offers a range of <a href="https://www.transkribus.org/public-models">publicly available models</a>, which can be used to transcribe medieval manuscripts in several different languages. Here are some of the most successful models for transcribing medieval manuscripts.
 
<h3>Latin and German Gothic Book Scripts</h3>
This model is designed for use with book scripts from the 13th-15th centuries and was trained with a variety of different documents in both Latin and German.
<a href="https://www.transkribus.org/model/latin-and-german-gothic-book-scripts">Try out this model</a>
 
<h3>Medieval Latin 13th-15th Century</h3>
Also suitable for documents from the 13th-15th centuries, this model was trained specifically for the transcription of legal documents in medieval Latin, including idiosyncratic abbreviations and random word hyphenations.
<a href="https://www.transkribus.org/model/medieval-latin">Try out this model</a>
 
<h3>Old Czech Handwriting (with spaces)</h3>
This model was trained on two Czech illuminated manuscripts, making it one of the more suitable models for manuscripts of this kind.
<a href="https://www.transkribus.org/model/old-czech-handwriting-with-spaces">Try out this model</a>
 
<h3>French and Latin Chancery Documents</h3>
Trained as part of the <a href="http://himanis.huma-num.fr/app//">HIMANIS project</a>, this model is designed for medieval chancery documents in French and Latin.
<a href="https://www.transkribus.org/model/french-and-latin-chancery-documents">Try out this model</a>
 
<h3>Latin and Dutch</h3>
For medieval legal records and minutes in Dutch, this model is ideal. It was trained by the <a href="https://www.huygens.knaw.nl/en/">Huygens Institute for History of the Netherlands</a>.
<a href="https://www.transkribus.org/model/latin-and-dutch">Try out this model</a>
 



<h2>Where can I find out more about transcribing medieval manuscripts with Transkribus?</h2>If you want to learn more about training text recognition models, Transkribus has plenty of resources to help. The <a href="https://help.transkribus.org">Transkribus Help Center</a> offers clear guides on using text recognition to automatically transcribe texts, as well as on all the other features available on the platform.Transkribus also hosts free webinars on topics such as text recognition, model training, and layout analysis. You can visit the <a href="https://www.transkribus.org/events">Events page</a> on the Transkribus website to sign up for these sessions—they’re a great chance to ask questions and learn directly from experts.

Gold, Gospels and Knights Fighting Snails: An Introduction To Illuminated Manuscripts

Gold, gospels and knights fighting snails: An introduction to illuminated manuscripts

<h2 class="wp-block-heading" id="h-ever-had-trouble-reading-someone-else-s-handwriting">Ever had trouble reading someone else’s handwriting?&nbsp;</h2>



 Well, it may reassure you to know that it’s not only humans that have this problem, but computers too. While computers have been able to recognise and transcribe printed text for decades, recognising handwritten texts has only been possible for the last few of years. The technology that does this is known as handwritten text recognition or HTR, and it is the foundation of the <a href="https://www.transkribus.org" target="_blank">Transkribus platform</a>.&nbsp;



<a href="https://www.transkribus.org/ai-text-recognition" target="_blank">Handwriting recognition</a> is a fascinating technology but it can be a bit complex to understand at the beginning. So if you are new to this new and unique technology, here is a quick introduction to what HTR is and what it does.



<h2 class="wp-block-heading"> </h2><h2 class="wp-block-heading">What is handwriting recognition or HTR?</h2>



 Handwriting recognition is a type of technology that can be used to “read” the writing in images of handwritten documents. Let’s say you wrote an essay by hand back when you were at school and you now want to have that essay as digital text on your computer. With the right HTR software, you could take a photo of the essay, run it through the software and get that same essay as digital text, which can be downloaded and shared as necessary. This is the basic principle of handwriting recognition. 



<figure class="wp-block-image size-large"><img decoding="async" width="1024" height="576" src="https://s3.eu-central-1.amazonaws.com/readcoop.cis.public-assets.prod/WP/uploads/2022/08/hwr-1-1024x576.jpg" alt="Handwriting Recogntion " class="wp-image-29726" srcset="https://s3.eu-central-1.amazonaws.com/readcoop.cis.public-assets.prod/WP/uploads/2022/08/hwr-1-1024x576.jpg 1024w, https://s3.eu-central-1.amazonaws.com/readcoop.cis.public-assets.prod/WP/uploads/2022/08/hwr-1-300x169.jpg 300w, https://s3.eu-central-1.amazonaws.com/readcoop.cis.public-assets.prod/WP/uploads/2022/08/hwr-1-768x432.jpg 768w, https://s3.eu-central-1.amazonaws.com/readcoop.cis.public-assets.prod/WP/uploads/2022/08/hwr-1-1536x864.jpg 1536w, https://s3.eu-central-1.amazonaws.com/readcoop.cis.public-assets.prod/WP/uploads/2022/08/hwr-1-18x10.jpg 18w, https://s3.eu-central-1.amazonaws.com/readcoop.cis.public-assets.prod/WP/uploads/2022/08/hwr-1-600x338.jpg 600w, https://s3.eu-central-1.amazonaws.com/readcoop.cis.public-assets.prod/WP/uploads/2022/08/hwr-1.jpg 1920w" sizes="(max-width: 1024px) 100vw, 1024px"></figure>



HTR technology turns handwritten documents and into digital text. © Transkribus We should also be clear on the terminology here. The kind of handwriting recognition described above is known as “offline handwriting recognition”. That is because it involves images of texts that has already been written. There is also “online handwriting recognition”. This is software that generates digital text from handwriting as you write it, usually with a tablet and stylus. As Transkribus was created to recognise the handwriting in historical documents that have already been written, it can only be used for offline handwriting recognition.



<h2 class="wp-block-heading"> </h2><h2 class="wp-block-heading">Why do computers find handwritten text harder to recognise than printed text?</h2>



 With printed texts, there is a finite number of fonts that can be used — 200,000 of them, in <a href="https://facts.net/fonts-facts/">fact</a>. And while this may seem like a lot, it does at least mean you can program software to read them all. This kind of technology is called <a href="https://readcoop.eu/insights/ocr-vs-htr/" target="_blank">Optical Character Recognition</a>, or OCR for short, and it has been used for decades with printed texts.



Handwriting is a different ball game. The 6.5 billion people on this planet who can write each have their own style of handwriting and no two styles are exactly the same. To make things more complicated, how someone writes on an official form might be very different to how they write in their own diary, for example. The infinite number of possibilities makes it also impossible to program software to read them all.&nbsp; 



<figure class="wp-block-video"><video autoplay="" controls="" loop="" preload="auto" src="https://s3.eu-central-1.amazonaws.com/readcoop.cis.public-assets.prod/WP/uploads/2022/08/hwr.mp4"></video><figcaption>Transkribus can read handwriting in any language or script. © Transkribus</figcaption></figure>



<h2 class="wp-block-heading"> </h2><h2 class="wp-block-heading">How does handwriting recognition work?</h2>



 So how does HTR technology recognise handwriting, if you can’t program it? Answer: HTR software doesn’t recognise handwriting, it learns to recognise handwriting. Over time, software like Transkribus uses AI and Deep Learning to learn how to read and transcribe different types of handwriting, just like a human would. To do this, you first need an AI model. This is like a giant digital mind that has learnt the shapes and characteristics of thousands of handwritten words, letters and even just symbols. Most importantly, the model can also make educated guesses about characters it hasn’t seen before. The software recognises the handwriting in the image and then uses the model’s knowledge to transcribe that handwriting into digital text. With the right model, you can achieve an almost perfect transcription at just the click of a button.&nbsp;



There are lots of <a href="https://readcoop.eu/transkribus/public-models/">public models</a> already available for Transkribus, covering different languages and types of handwriting such as <a href="https://www.transkribus.org/model/german-kurrent-and-sutterlin-17th-20th-century" target="_blank">German Kurrent</a>, or <a href="https://www.transkribus.org/model/english-handwriting-18th-19th-century-2" target="_blank">English handwriting from the 18th and 19th centuries</a>. But you can also <a href="https://help.transkribus.org/training-text-recognition-models" target="_blank">create your own model</a>. This is done by manually transcribing a certain number of documents, to train the “digital mind” to recognise the specific handwriting in your documents. You can then use this custom model to automatically transcribe the rest of your documents. 



<h2 class="wp-block-heading"> </h2><h2 class="wp-block-heading">What do people use HTR technology for?</h2>



 There are many jobs that involve being able to quickly read handwritten documents. For example, <a href="https://www.transkribus.org/success-story/transcribing-the-personal-album-of-zoila-aurora-caceres" target="_blank">historians</a> and other <a href="https://www.transkribus.org/success-story/an-gaodhal" target="_blank">researchers</a> use historical handwritten documents to learn about the past. They often want to create digital versions of the documents, so that they can more easily analyse the documents and search the whole collection for certain words or terms. HTR technology is perfect for this.



<figure class="wp-block-image size-large"><img decoding="async" width="1024" height="772" src="https://s3.eu-central-1.amazonaws.com/readcoop.cis.public-assets.prod/WP/uploads/2022/08/rses-1024x772.jpg" alt="" class="wp-image-29728" srcset="https://s3.eu-central-1.amazonaws.com/readcoop.cis.public-assets.prod/WP/uploads/2022/08/rses-1024x772.jpg 1024w, https://s3.eu-central-1.amazonaws.com/readcoop.cis.public-assets.prod/WP/uploads/2022/08/rses-300x226.jpg 300w, https://s3.eu-central-1.amazonaws.com/readcoop.cis.public-assets.prod/WP/uploads/2022/08/rses-768x579.jpg 768w, https://s3.eu-central-1.amazonaws.com/readcoop.cis.public-assets.prod/WP/uploads/2022/08/rses-1536x1158.jpg 1536w, https://s3.eu-central-1.amazonaws.com/readcoop.cis.public-assets.prod/WP/uploads/2022/08/rses-2048x1544.jpg 2048w, https://s3.eu-central-1.amazonaws.com/readcoop.cis.public-assets.prod/WP/uploads/2022/08/rses-16x12.jpg 16w, https://s3.eu-central-1.amazonaws.com/readcoop.cis.public-assets.prod/WP/uploads/2022/08/rses-600x452.jpg 600w" sizes="(max-width: 1024px) 100vw, 1024px"><figcaption>You can publish your documents online via <a href="https://www.transkribus.org/de/sites" target="_blank">Transkribus Sites</a>. © Transkribus</figcaption></figure>



 <a href="https://www.transkribus.org/success-story/national-archives-of-the-netherlands" target="_blank">Archives</a> and <a href="https://www.transkribus.org/success-story/norhand" target="_blank">libraries</a> are also places where HTR technology can be useful. Nowadays, archivists and librarians aim to offer digital versions of as many books and documents as possible. But manually transcribing the hundreds and thousands of volumes in a standard archive would take forever. Automatic transcription with HTR is a much quicker alternative, and makes it easier to publish all the documents online too.



But there are many other people who use HTR technology to make their lives easier: <a href="https://www.transkribus.org/success-story/the-pandetta-project-fondazione-banco-di-napoli" target="_blank">companies wanting to digitise their old files</a>, <a href="https://www.transkribus.org/blog/read-coop-and-the-europeana-foundation-join-forces" target="_blank">developers wanting to incorporate handwriting recognition into custom software</a>, <a href="https://www.transkribus.org/success-story/trug-und-schein" target="_blank">hobby genealogists who want to read their family’s old records</a>, the list is endless! 



<h2 class="wp-block-heading"> </h2><h2 class="wp-block-heading">Can I see an example of handwriting recognition in action?</h2>



 If you want to see how handwriting recognition works, then simply upload a scanned image of handwritten text to the widget below, and see the technology in action. 



<iframe width="100%" height="600" src="https://transkribus.ai/?embed=1" frameborder="0"></iframe>

What is handwriting recognition and how does it work?

The latest version of Transkribus Lite is here and brings a number of new features. Here are the most important new features:&nbsp;



<h2 class="wp-block-heading" id="h-smart-search">Smart Search</h2>



From now on, you can simply tick a checkbox when you are starting a text recognition job to enhance the search experience in your documents. With Smart Search, Transkribus can be configured to store a large number of the possible words for each place in the document, and to make these available for searching. Thus it is possible to find search terms that would not be found with a regular full-text search that stores only one candidate per word. This feature comes in handy especially when searching for named entities, such as place and person names, because these are typically difficult to &#8220;learn&#8221; for Transkribus as they are relatively rare compared to &#8220;normal&#8221; words. The possible alternative words are identified using AI and thus selected in a smart way. If you choose to run text recognition with Smart Search switched on, this improved search experience will be available both in Transkribus itself and on your <a href="https://readcoop.eu/readsearch/">read&amp;search</a> website, should you have one. 



<figure class="wp-block-image size-large"><img decoding="async" width="1024" height="583" src="https://s3.eu-central-1.amazonaws.com/readcoop.cis.public-assets.prod/WP/uploads/2022/07/smart-search-1024x583.jpg" alt="" class="wp-image-29579" srcset="https://s3.eu-central-1.amazonaws.com/readcoop.cis.public-assets.prod/WP/uploads/2022/07/smart-search-1024x583.jpg 1024w, https://s3.eu-central-1.amazonaws.com/readcoop.cis.public-assets.prod/WP/uploads/2022/07/smart-search-300x171.jpg 300w, https://s3.eu-central-1.amazonaws.com/readcoop.cis.public-assets.prod/WP/uploads/2022/07/smart-search-768x438.jpg 768w, https://s3.eu-central-1.amazonaws.com/readcoop.cis.public-assets.prod/WP/uploads/2022/07/smart-search-1536x875.jpg 1536w, https://s3.eu-central-1.amazonaws.com/readcoop.cis.public-assets.prod/WP/uploads/2022/07/smart-search-2048x1167.jpg 2048w, https://s3.eu-central-1.amazonaws.com/readcoop.cis.public-assets.prod/WP/uploads/2022/07/smart-search-18x10.jpg 18w, https://s3.eu-central-1.amazonaws.com/readcoop.cis.public-assets.prod/WP/uploads/2022/07/smart-search-600x342.jpg 600w" sizes="(max-width: 1024px) 100vw, 1024px" /><figcaption>Smart Search at the text recognition dialogue</figcaption></figure>



<h2 class="wp-block-heading">The tag manager</h2>



With the new tag manager you can easily manage, view and search the tags in your collections. The edit mode lets you change existing tags, remove existing tags, and add new tags to your collection. The list mode displays all tags that you have used in your collection and the number of times that they have been used. With the tag search feature you can search your tags with ease. 



<figure class="wp-block-image size-large"><img decoding="async" width="1024" height="478" src="https://s3.eu-central-1.amazonaws.com/readcoop.cis.public-assets.prod/WP/uploads/2022/07/tag-manager-1024x478.jpg" alt="" class="wp-image-29580" srcset="https://s3.eu-central-1.amazonaws.com/readcoop.cis.public-assets.prod/WP/uploads/2022/07/tag-manager-1024x478.jpg 1024w, https://s3.eu-central-1.amazonaws.com/readcoop.cis.public-assets.prod/WP/uploads/2022/07/tag-manager-300x140.jpg 300w, https://s3.eu-central-1.amazonaws.com/readcoop.cis.public-assets.prod/WP/uploads/2022/07/tag-manager-768x359.jpg 768w, https://s3.eu-central-1.amazonaws.com/readcoop.cis.public-assets.prod/WP/uploads/2022/07/tag-manager-1536x717.jpg 1536w, https://s3.eu-central-1.amazonaws.com/readcoop.cis.public-assets.prod/WP/uploads/2022/07/tag-manager-2048x956.jpg 2048w, https://s3.eu-central-1.amazonaws.com/readcoop.cis.public-assets.prod/WP/uploads/2022/07/tag-manager-18x8.jpg 18w, https://s3.eu-central-1.amazonaws.com/readcoop.cis.public-assets.prod/WP/uploads/2022/07/tag-manager-600x280.jpg 600w" sizes="(max-width: 1024px) 100vw, 1024px" /><figcaption>The new tag manager</figcaption></figure>







<h2 class="wp-block-heading">Add relations to the layout of your documents</h2>



The layout mode of the document editor now allows you to create relations between layout elements. This way, you can connect related layout elements for further use cases. The relational data will also be incorporated into the XML of the document.



<figure class="wp-block-image size-large"><img decoding="async" width="1024" height="477" src="https://s3.eu-central-1.amazonaws.com/readcoop.cis.public-assets.prod/WP/uploads/2022/07/layout-manager-1024x477.jpg" alt="" class="wp-image-29581" srcset="https://s3.eu-central-1.amazonaws.com/readcoop.cis.public-assets.prod/WP/uploads/2022/07/layout-manager-1024x477.jpg 1024w, https://s3.eu-central-1.amazonaws.com/readcoop.cis.public-assets.prod/WP/uploads/2022/07/layout-manager-300x140.jpg 300w, https://s3.eu-central-1.amazonaws.com/readcoop.cis.public-assets.prod/WP/uploads/2022/07/layout-manager-768x357.jpg 768w, https://s3.eu-central-1.amazonaws.com/readcoop.cis.public-assets.prod/WP/uploads/2022/07/layout-manager-1536x715.jpg 1536w, https://s3.eu-central-1.amazonaws.com/readcoop.cis.public-assets.prod/WP/uploads/2022/07/layout-manager-2048x953.jpg 2048w, https://s3.eu-central-1.amazonaws.com/readcoop.cis.public-assets.prod/WP/uploads/2022/07/layout-manager-18x8.jpg 18w, https://s3.eu-central-1.amazonaws.com/readcoop.cis.public-assets.prod/WP/uploads/2022/07/layout-manager-600x279.jpg 600w" sizes="(max-width: 1024px) 100vw, 1024px" /><figcaption>Adding relations in the layout editor</figcaption></figure>



<h2 class="wp-block-heading">Improved user interface</h2>



With this new release we have also put a strong focus on improving the user experience by making the user interface more simple and intuitive. Now you can easily distinguish between documents and single pages as the design of the document as well as page overview has been adapted to make the difference clear. Some additional changes to the sidebar, the top navigation as well as to a number of other, smaller elements will make working with Transkribus Lite more intuitive and easy. 



<figure class="wp-block-image size-large"><img decoding="async" width="1024" height="576" src="https://s3.eu-central-1.amazonaws.com/readcoop.cis.public-assets.prod/WP/uploads/2022/07/new-interface-1024x576.jpg" alt="" class="wp-image-29582" srcset="https://s3.eu-central-1.amazonaws.com/readcoop.cis.public-assets.prod/WP/uploads/2022/07/new-interface-1024x576.jpg 1024w, https://s3.eu-central-1.amazonaws.com/readcoop.cis.public-assets.prod/WP/uploads/2022/07/new-interface-300x169.jpg 300w, https://s3.eu-central-1.amazonaws.com/readcoop.cis.public-assets.prod/WP/uploads/2022/07/new-interface-768x432.jpg 768w, https://s3.eu-central-1.amazonaws.com/readcoop.cis.public-assets.prod/WP/uploads/2022/07/new-interface-1536x864.jpg 1536w, https://s3.eu-central-1.amazonaws.com/readcoop.cis.public-assets.prod/WP/uploads/2022/07/new-interface-18x10.jpg 18w, https://s3.eu-central-1.amazonaws.com/readcoop.cis.public-assets.prod/WP/uploads/2022/07/new-interface-600x338.jpg 600w, https://s3.eu-central-1.amazonaws.com/readcoop.cis.public-assets.prod/WP/uploads/2022/07/new-interface.jpg 1920w" sizes="(max-width: 1024px) 100vw, 1024px" /><figcaption>New layout of Documents and Pages</figcaption></figure>

Transkribus Lite 2.3. – New Features: Smart Search, Tag Manager, and Layout Relations

The latest addition to the long list of Transkribus public models comes from the <a href="https://www.arkivverket.no/en">National Archives of Norway</a>. Thanks to their efforts, not one, not two, but three new models for Norwegian (and also Danish) are now publicly available.Two of the models are for typewritten material: one for <a href="https://readcoop.eu/model/danish-norwegian-typewritten-transcripts-1760-1842/">Danish-Norwegian typewritten transcripts</a> and the other for <a href="https://readcoop.eu/model/norwegian-typewriting-1940-1945/">Norwegian typewriting</a> from 1940-1945. Both models have impressive CERs of less than 1%.&nbsp;The third of the three models, one could say the peak of the trident, is a handwriting model for <a href="https://readcoop.eu/model/danish-norwegian-handwriting-1874-1891/">Danish &amp; Norwegian Handwriting</a> from 1874–1891. This model is based on partition protocols for Sunnhordland, Norway. This Norwegian trident is now available to anyone—you can try them out now on our website.

The Norwegian Trident &#8211; Three New Public Models For Norwegian (&#038; Danish)

By Joe Nockels, University of Edinburgh



As part of his PhD research at the University of Edinburgh and National Library of Scotland (NLS), Joe Nockels underwent a systematic review of how Trankribus was being mentioned in research from 2015-2020. A total of 381 papers were gathered from Google Scholar, Scopus, and Web of Science. Here are a few of the findings:



Published research mentioning Transkribus is international and growing.



There was a 235% increase in Transkribus research between 2017 and 2018, returning 39 and 92 works respectively. 2019 saw another increase in materials mentioning Transkribus, rising to 112. With the analysis ending in October 2020, understandably fewer works were, although the amount of research was still sizeable with 99 materials. Transkribus&#8217;s rise in mentions eludes to a shift in the collection and curatorial landscapes, with memory institutions opting for the digitisation and recognition of their materials en masse (Chassanoff 2013; Duff et al. 2004). It is a reasonable assertion to think that this has only increased during and after the Covid-19 pandemic, which inadvertently provided a window for archives and libraries to prioritise digital projects, delivering services through multiple channels for those without access to buildings.



English was the most common language of Transkribus research (67.98%, n = 259). Transkribus research appears more multilingual than the general content in Scopus; where 88.4% of results, 77% of art and humanities materials, appeared in English in 2013 (Van Weijen 2013). More recent studies have shown that the prevalence of Standard English is on the rise in scholarly research, irrespective of field, estimating that 98% of publications in science are written in English, causing researchers from English as a Foreign Language (EFL) countries to sound the alarm that their contributions are being inhibited (Flowerdew 2013; Ramirez-Castaneda 2020). We should remain mindful of this trend, although English materials mentioning Transkribus sits alongside works in Dutch, Spanish; Swedish, Bosnian, Russian, Norwegian, Polish, Italian, Croatian, Hungarian, Czech and even Maori, showcasing the diversity of those researching Transkribus, as well the broad array of their research.



<figure class="wp-block-image size-large"><img decoding="async" width="1024" height="628" src="https://s3.eu-central-1.amazonaws.com/readcoop.cis.public-assets.prod/WP/uploads/2022/06/map-transkribus-in-research-1024x628.jpg" alt="" class="wp-image-29368" srcset="https://s3.eu-central-1.amazonaws.com/readcoop.cis.public-assets.prod/WP/uploads/2022/06/map-transkribus-in-research-1024x628.jpg 1024w, https://s3.eu-central-1.amazonaws.com/readcoop.cis.public-assets.prod/WP/uploads/2022/06/map-transkribus-in-research-300x184.jpg 300w, https://s3.eu-central-1.amazonaws.com/readcoop.cis.public-assets.prod/WP/uploads/2022/06/map-transkribus-in-research-768x471.jpg 768w, https://s3.eu-central-1.amazonaws.com/readcoop.cis.public-assets.prod/WP/uploads/2022/06/map-transkribus-in-research-1536x943.jpg 1536w, https://s3.eu-central-1.amazonaws.com/readcoop.cis.public-assets.prod/WP/uploads/2022/06/map-transkribus-in-research-2048x1257.jpg 2048w, https://s3.eu-central-1.amazonaws.com/readcoop.cis.public-assets.prod/WP/uploads/2022/06/map-transkribus-in-research-18x12.jpg 18w, https://s3.eu-central-1.amazonaws.com/readcoop.cis.public-assets.prod/WP/uploads/2022/06/map-transkribus-in-research-600x368.jpg 600w" sizes="(max-width: 1024px) 100vw, 1024px" /><figcaption>Transkribus research (2015-2020) plotted using Digimap’s roam feature. Location markers identify the lead researchers’ affiliation. The Innsbruck server is located with a yellow pin.</figcaption></figure>







Plotting returned works geographically, we can see that materials mentioning Transkribus gravitate around the servers held at Innsbruck, with German institutions the most represented. That said, research avenues have opened up in West Asia particularly.



Members of the READ-COOP play a large role in publishing research using and mentioning Transkribus.



118 works (30.97%) came from READ-COOP institutions. This highlights that the COOP is becoming a strong space for like-minded institutions to feature Transkribus, presenting results and increasing its capabilities as a tool.



<figure class="wp-block-image size-large"><img decoding="async" width="1024" height="551" src="https://s3.eu-central-1.amazonaws.com/readcoop.cis.public-assets.prod/WP/uploads/2022/06/transkribus-in-research-1024x551.jpg" alt="" class="wp-image-29369" srcset="https://s3.eu-central-1.amazonaws.com/readcoop.cis.public-assets.prod/WP/uploads/2022/06/transkribus-in-research-1024x551.jpg 1024w, https://s3.eu-central-1.amazonaws.com/readcoop.cis.public-assets.prod/WP/uploads/2022/06/transkribus-in-research-300x161.jpg 300w, https://s3.eu-central-1.amazonaws.com/readcoop.cis.public-assets.prod/WP/uploads/2022/06/transkribus-in-research-768x413.jpg 768w, https://s3.eu-central-1.amazonaws.com/readcoop.cis.public-assets.prod/WP/uploads/2022/06/transkribus-in-research-18x10.jpg 18w, https://s3.eu-central-1.amazonaws.com/readcoop.cis.public-assets.prod/WP/uploads/2022/06/transkribus-in-research-600x323.jpg 600w, https://s3.eu-central-1.amazonaws.com/readcoop.cis.public-assets.prod/WP/uploads/2022/06/transkribus-in-research.jpg 1382w" sizes="(max-width: 1024px) 100vw, 1024px" /><figcaption>Chart highlighting the research output of institutional READ-COOP members by number of publications.</figcaption></figure>







Transkribus is mainly mentioned in journal articles (42.87%, n = 163), conference papers (21.00%, n = 80) and policy documents (8.14%, n = 31). Although research using and mentioning the software appeared in greyer materials like book sections, undergraduate and postgraduate theses, reports, presentations, blog posts, magazines and video recordings. Out of the entire dataset, 71.13% of works were accessible online under an open copyright license (n= 271), replicating to an extent the atmosphere of collegiality and cooperation seen with the establishment of the shareholding model of the READ-COOP. This is a positive sign, enabling greater research to be done through the sharing of publications, models and methodologies.



Transkribus features primarily in archival and library science publications, while a broad range of disciplines are using HTR to a lesser extent such as: history, computer science, citizen science, law and education. This shows the wide applicability of the HTR tool, providing a host of useful tools at researchers&#8217; disposal: improving collection descriptions, information retrieval and the recognition of historical documents. Transkribus is useful regardless of users’ concept of data-driven methods. A librarian may want to use HTR not to produce full transcriptions but to keyword spot metadata across collections, improving access to historical materials. By contrast, individual researchers tend to use Transkribus to produce rich data that is ‘replete with enough specifics that they may operationalize that data in pursuit of their research goals’. In terms of the domain of computer science, research into Transkribus is more dependent on gaining predictable and regularised results (such as a certain level of character/word error rate) (Lincoln 2017, p 30). Transkribus is flexible enough to meet all these specific needs.



Since 2017 especially, research into Transkribus has been undertaken in a variety of fields. While archival science, information science, and computer science remain the dominant disciplines, we found that work has been published across the arts, humanities and social sciences (AHSS). Fields that were represented in our corpus included religious studies, publishing, history, theatre studies, philosophy, management science, and medieval studies.



Taking all these findings into account, Transkribus appears to be as Thylstrup (2019) describes a ‘bottom-up’ mass digitization movement, made up of hundreds of simultaneous projects driven by motivated researchers. This provides a firm base for the READ-COOP, increasing the accuracy of Transkribus by supplying ground truth data and trained models. As Transkribus grows, more research will inevitably be produced and new rhythms will emerge in the approaches of researchers. Carrying out subsequent analyses of the literature will allow this to be tracked and understood. Through such a structure, the evidential value of user experiences with HTR can be harnessed and collected, helping to develop Transkribus in a sustainable and useful manner.



For the full paper see: <a href="https://link.springer.com/article/10.1007/s10502-022-09397-0">https://link.springer.com/article/10.1007/s10502-022-09397-0</a>



Chassanof A (2013) Historians and the use of primary source materials in the digital age. Am Arch 76:458–480. https://doi.org/10.17723/aarc.76.2.lh76217m2m376n28



Duff W, Craig B, Cherry J (2004) Historians’ use of archival sources: promises and pitfalls of the digital age. Public Hist 26:1–10



Flowerdew J (2013) Some thoughts on English for Research Publication Purposes (ERPP) and related issues. Cambridge University Press, Cambridge



Lincoln M (2017) Ways of forgetting: the librarian, the historian, and the machine. In: Padilla T, Allen L, Frost H, Potvin S, Russey RE, Varner S (eds) Always already computational: library collections as data. Institute of Memory and Library Services, National Forum Positional Statements, pp 20–30. https://collectionsasdata.github.io. Accessed 20 Nov 2020



Ramirez-Castaneda V (2020) Disadvantages in preparing and publishing scientific papers caused by the dominance of the English language in Science: the case of Colombian researchers in biological sciences. Paper presented at PLoS One, Kyoto, 16 Sept 2020. https://doi.org/10.1371/journal.pone. 0238372



Thylstrup NB (2019) The politics of mass digitization. MIT Press, Cambridge



van Weijen D (2013) Publication languages in the arts &amp; humanities. Res Trends 32:1–10

Understanding the application of Transkribus in research

We are excited to launch our new “<a href="https://zoom.us/meeting/register/tJUvdeyvrj4uHNTSFv7JwVk6A0veTORATUaL">Ask Us Anything</a>” webinar series, where attendees can ask us about anything related to Transkribus, handwritten text recognition, transcribing and related topics. This webinar series provides an opportunity for dialogue and feedback that will help us to improve Transkribus and other related tools. We hope you&#8217;ll join us for this series and take advantage of this opportunity to ask us anything!



Join us on the 29th of June at 5 p.m. (CEST) for our Ask Us Anything webinar. This is your chance to get all your questions answered. Register now and we&#8217;ll see you on the 29th of June.



You can register for the webinar <a href="https://zoom.us/meeting/register/tJUvdeyvrj4uHNTSFv7JwVk6A0veTORATUaL">here</a>.

New Webinar: Ask Us Anything About Transkribus

<h2 class="wp-block-heading" id="h-the-new-sharing-feature">The new sharing feature</h2>



It might be a small button for a screen, but it is a great function for your work: The new sharing feature in <a href="https://lite.transkribus.eu">Transkribus Lite</a>. Want to quickly share a page you are working on with others? You no longer need to screenshot it, you can share it instead. With just two clicks, you can now share a page from Transkribus Lite with anyone, even if they’re not registered with Transkribus. The new sharing feature also gives you full control over your shared page so you can make it private again at any time. Try out the new feature and let us know what you think!


<div class="wp-block-image">
<figure class="aligncenter size-full"><img decoding="async" width="602" height="148" src="https://s3.eu-central-1.amazonaws.com/readcoop.cis.public-assets.prod/WP/uploads/2022/06/lite22_1.jpg" alt="" class="wp-image-29216" srcset="https://s3.eu-central-1.amazonaws.com/readcoop.cis.public-assets.prod/WP/uploads/2022/06/lite22_1.jpg 602w, https://s3.eu-central-1.amazonaws.com/readcoop.cis.public-assets.prod/WP/uploads/2022/06/lite22_1-300x74.jpg 300w, https://s3.eu-central-1.amazonaws.com/readcoop.cis.public-assets.prod/WP/uploads/2022/06/lite22_1-18x4.jpg 18w, https://s3.eu-central-1.amazonaws.com/readcoop.cis.public-assets.prod/WP/uploads/2022/06/lite22_1-600x148.jpg 600w" sizes="(max-width: 602px) 100vw, 602px" /></figure></div>


A small button with a big impact: The new sharing button



<h2 class="wp-block-heading">Improved navigation bar</h2>



With this new feature we have also improved the navigation bar in the single page view. Now you can quickly navigate between the three different modes (view-, text editing and layout editing mode). This new streamlined look&nbsp; makes it easier to focus on what is really important: Working on your documents.&nbsp;



<h2 class="wp-block-heading">Be in full control of your metadata</h2>



From now on you can also edit all the metadata of your documents and collections. Just open the edit mode via the context menu with the three dots and you can modify all the additional information for your documents and collections.


<div class="wp-block-image">
<figure class="aligncenter size-full"><img decoding="async" width="602" height="307" src="https://s3.eu-central-1.amazonaws.com/readcoop.cis.public-assets.prod/WP/uploads/2022/06/lite22_2.jpg" alt="" class="wp-image-29217" srcset="https://s3.eu-central-1.amazonaws.com/readcoop.cis.public-assets.prod/WP/uploads/2022/06/lite22_2.jpg 602w, https://s3.eu-central-1.amazonaws.com/readcoop.cis.public-assets.prod/WP/uploads/2022/06/lite22_2-300x153.jpg 300w, https://s3.eu-central-1.amazonaws.com/readcoop.cis.public-assets.prod/WP/uploads/2022/06/lite22_2-18x9.jpg 18w, https://s3.eu-central-1.amazonaws.com/readcoop.cis.public-assets.prod/WP/uploads/2022/06/lite22_2-600x306.jpg 600w" sizes="(max-width: 602px) 100vw, 602px" /></figure></div>


<h2 class="wp-block-heading">Use the new table view to master large collections</h2>



This new table view is ideal for collections containing large amounts of documents With a simple click, you can switch between the cards and table overview of your documents, allowing you to view the collection in the way that is easiest for you


<div class="wp-block-image">
<figure class="aligncenter size-full is-resized"><img decoding="async" src="https://s3.eu-central-1.amazonaws.com/readcoop.cis.public-assets.prod/WP/uploads/2022/06/lite22_3.jpg" alt="" class="wp-image-29218" width="602" height="211" srcset="https://s3.eu-central-1.amazonaws.com/readcoop.cis.public-assets.prod/WP/uploads/2022/06/lite22_3.jpg 602w, https://s3.eu-central-1.amazonaws.com/readcoop.cis.public-assets.prod/WP/uploads/2022/06/lite22_3-300x105.jpg 300w, https://s3.eu-central-1.amazonaws.com/readcoop.cis.public-assets.prod/WP/uploads/2022/06/lite22_3-18x6.jpg 18w, https://s3.eu-central-1.amazonaws.com/readcoop.cis.public-assets.prod/WP/uploads/2022/06/lite22_3-600x210.jpg 600w" sizes="(max-width: 602px) 100vw, 602px" /></figure></div>


<h2 class="wp-block-heading">New: Work with structure types</h2>



If you want to work with structure types, Transkribus Lite has you covered. Using structural tags, you can now mark up the entire structure of your documents. It is also possible to&nbsp; set up and modify your structure types on a collection level in the collection overview, and assign different structure types to text regions simply by right clicking and assigning them in the layout editor.&nbsp;


<div class="wp-block-image">
<figure class="aligncenter size-full is-resized"><img decoding="async" src="https://s3.eu-central-1.amazonaws.com/readcoop.cis.public-assets.prod/WP/uploads/2022/06/lite22_4.jpg" alt="" class="wp-image-29219" width="602" height="163" srcset="https://s3.eu-central-1.amazonaws.com/readcoop.cis.public-assets.prod/WP/uploads/2022/06/lite22_4.jpg 602w, https://s3.eu-central-1.amazonaws.com/readcoop.cis.public-assets.prod/WP/uploads/2022/06/lite22_4-300x81.jpg 300w, https://s3.eu-central-1.amazonaws.com/readcoop.cis.public-assets.prod/WP/uploads/2022/06/lite22_4-18x5.jpg 18w, https://s3.eu-central-1.amazonaws.com/readcoop.cis.public-assets.prod/WP/uploads/2022/06/lite22_4-600x162.jpg 600w" sizes="(max-width: 602px) 100vw, 602px" /></figure></div>

<div class="wp-block-image">
<figure class="aligncenter size-full is-resized"><img decoding="async" src="https://s3.eu-central-1.amazonaws.com/readcoop.cis.public-assets.prod/WP/uploads/2022/06/lite22_5.jpg" alt="" class="wp-image-29220" width="602" height="211" srcset="https://s3.eu-central-1.amazonaws.com/readcoop.cis.public-assets.prod/WP/uploads/2022/06/lite22_5.jpg 602w, https://s3.eu-central-1.amazonaws.com/readcoop.cis.public-assets.prod/WP/uploads/2022/06/lite22_5-300x105.jpg 300w, https://s3.eu-central-1.amazonaws.com/readcoop.cis.public-assets.prod/WP/uploads/2022/06/lite22_5-18x6.jpg 18w, https://s3.eu-central-1.amazonaws.com/readcoop.cis.public-assets.prod/WP/uploads/2022/06/lite22_5-600x210.jpg 600w" sizes="(max-width: 602px) 100vw, 602px" /></figure></div>

Transkribus Lite 2.2.: Sharing your work has just got easier 

Many people are surprised to learn that Transkribus is not run by a regular company, but by a cooperative: READ-COOP. That means that it is our members, not investors, that dictate how the organisation is run — after all, they are the people who use our technology every day to make the past accessible for everyone. Our READ-COOP members range from individuals with a passion for historical manuscripts to universities, research institutes, and libraries. However, in celebration of <a href="https://www.ica.org/en/events-and-resources/about-international-archives-day-and-international-archives-week">International Archives Day</a>, we would like to focus today on one particular section of our membership: archives. We are lucky to have several amazing archives as members of our cooperative, so let’s take a quick tour abroad to find out about the fascinating projects these archives carry out with Transkribus.



<h3 class="wp-block-heading" id="h-the-nordic-and-baltic-countries">The Nordic and Baltic Countries</h3>



Birth registers, military records, court proceedings: national archives are literally mines of historical information about a country. Each archive is run by a team of people responsible for preserving these enlightening documents and making them accessible to the public. Nowadays, that means digitising them and creating an online database, and many archives have achieved this using READ-COOP technology.



National archives in the Nordic and Baltic countries that are members of READ-COOP include the <a href="https://kansallisarkisto.fi/en/frontpage">Kansallisarkisto </a>in Finland, the <a href="https://www.arkivverket.no/en">Arkivverket</a> in Norway, and the <a href="https://dkarkiver.dk/">Organisation of Danish Archives</a>. Part of our members are also the <a href="https://riksarkivet.se/startpage">Riksarkivet</a> in Sweden, as well as the <a href="https://www.ra.ee/en/">Rahvusarhiiv</a> neighbouring Estonia. All of these archives used <a href="https://readcoop.eu/transkribus">Transkribus</a> to help digitise their vast collections of documents, and three — Sweden,Finland and Estonia — have also made their collections public via the <a href="https://readcoop.eu/readsearch/">read&amp;search</a> platform. That way, researchers and the public can search the entire archive and quickly find the information they need.



As part of the citizen science project <a href="https://www.retrodigitalisering.dk/about/">RETRO</a>, the Organisation of Danish Archives has also transcribed a large amount of material, mostly municipal documents. They have now made their text recognition model of <a href="https://readcoop.eu/model/danish-gothic-handwriting/">19th-century Danish Gothic handwriting</a> available to the public. Also in the Nordics, there is the archive of the <a href="https://www.sls.fi/en">Society of Swedish Literature in Finland</a>, who preserve and promote Swedish-Finnish culture.



<h3 class="wp-block-heading" id="h-belgium-the-netherlands-and-luxembourg">Belgium, the Netherlands and Luxembourg</h3>



In the Netherlands, it is common that individual cities and regions run archives of information about the area and the people who have lived there. Many of these city archives are members of READ-COOP, including <a href="https://hetutrechtsarchief.nl/english">Utrecht</a>, <a href="https://www.amsterdam.nl/stadsarchief/">Amsterdam</a>, <a href="https://noord-hollandsarchief.nl/">North Holland</a> and <a href="https://www.groningerarchieven.nl/">Groningen</a>. By converting their manuscripts into readable and searchable modern text with Transkribus, these Dutch archives have uncovered some fascinating new pieces of history, not to mention a <a href="https://readcoop.eu/success-stories/amsterdam-notary-archives-out-of-the-dark/">Rembrandt painting</a> no one even knew had existed!&nbsp;



Among the more recent members is the <a href="https://arch.arch.be/index.php?l=en">State Archives of Belgium</a>. Consisting of the National Archives in Brussels and many archival respositories throughout the country, the State Archives of Belgium decided to join READ-COOP with the common mission to make archives more accessible. Luxembourg may be one of the smallest countries in Europe, but they are also home to an impressively large archive: the <a href="https://anlux.public.lu/de.html">Archives Nationales de Luxembourg</a>.



<h3 class="wp-block-heading" id="h-germany-austria-and-switzerland">Germany, Austria, and Switzerland</h3>



The <a href="https://www.stockach.de/tourismus-kultur/de/kulturzentrum/stadtarchiv-1">City Archive of Stockach</a> in Germany also decided to become part of the READ-COOP and to work with Transkribus to recognise handwritten as well as printed documents. In the German-speaking world, however, it is not just governments or municipalities that run archives, but churches too. The diocese of <a href="https://www.bistum-passau.de/bistum-pfarreien/geschichte-archiv">Passau</a> in Germany and <a href="https://www.dasp.at/">St. Pölten</a> in Austria are both READ-COOP members and have extensive archives of church records dating back to the 16th century. In neighbouring Switzerland, there are two archives that are members of READ-COOP: the <a href="https://www.zh.ch/de/direktion-der-justiz-und-des-innern/staatsarchiv.html#342541443">State Archive in Zürich</a> and the <a href="https://www.stadtzug.ch/stadtarchiv">City Archive in Zug</a>. Along with preserving and digitising historical documents, these two archives also run a host of educational events and tours, including <a href="https://www.zh.ch/de/news-uebersicht/mitteilungen/2022/politik-staat/staatsarchiv/-transkribus--fuer-archive0.html">Transkribus workshops</a>.



<h3 class="wp-block-heading" id="h-italy">Italy</h3>



Our southernmost member in Europe is the <a href="http://www.fondazionebanconapoli.it/">Fondazione Banco di Napoli</a> in Naples, Italy. This vast archive holds financial records from various Neapolitan banks from the 16th century onwards, documenting the city’s rise as one of the most important trading posts in Europe. As part of the <a href="https://readcoop.eu/success-stories/the-pandetta-project-fondazione-banco-di-napoli/">Pandetta project</a>, the Fondazione is currently digitising their archives — a monumental task given its size. 190,000 individual records have already been scanned and digitally transcribed, covering the period from 1586 to 1599. So only another 400 years of history to go!



<h3 class="wp-block-heading" id="h-argentina">Argentina</h3>



And last but not certainly not least, our final member is the <a href="https://www.fundacionbyb.org/fbb-en">Fundación Bunge y Born</a>, a non-profit organisation based in Buenos Aires, Argentina. One of the areas of interest of the Fundación Bunge y Born is historical documentation and the preservation of cultural heritage. Together with Transkribus and the use of text recognition and digitisation technologies, they are working on a more digital and accessible historical archive.



We’d like to take this opportunity to thank all our members for your continued support and collaboration on our mission to make archives digitally available to everyone!







<h2 class="wp-block-heading" id="h-want-to-become-a-member-of-read-coop">Want to Become a Member of READ-COOP?</h2>



If you are an archivist, a researcher, a historian or anyone else with an interest in digitising and publishing historical documents for the greater good of society, then you might want to consider becoming a member of READ-COOP.



Our cooperative is dedicated to developing technologies that seamlessly open up historical manuscripts and documents. Technologies already in existence include our Transkribus software, <a href="https://scantent.eu/">ScanTent</a> photo aid, and the Transkribus Sites search platform and we constantly have new ideas and innovations in the pipeline.



READ-COOP is a registered European cooperative that is governed by our members, not finance-focused investors. Our motto is “purpose before profit”. Being a member not only gives you a say in the future of our technologies, but also allows you to use Transkribus at a discounted price and gives you access to privileged information about the organisation.



For more information and to become a member, please visit our <a href="https://readcoop.eu/join/">member’s page</a>.

Archives Who Are Members Of READ-COOP

The <a href="https://readcoop.eu/tuc22/">Trankribus User Conference</a> is back! This year, we are once again hosting a live event in Innsbruck, packed full of workshops and presentations, not to mention the chance to meet Transkribus users from all over the globe. 



Like every year, you have the chance to present at TUC22 on a topic of your choice. Past topics have included crowd-sourcing, structure recognition, research on various periods, as well as various new Transkribus features. And if you want to get involved in this year’s TUC, then it’s not too late! We have just extended the deadline for contributions to 15 June, so there is still time to send us your ideas. But we recommend reading this blog post first, to find out everything you need to know about contributing to the TUC22.



<h2 class="wp-block-heading" id="h-what-is-the-tuc">What is the TUC?</h2>



The Transkribus User Conference is our annual meetup for the global Transkribus community.&nbsp; With a two-day programme of talks and workshops, TUC is an excellent way to expand your Transkribus skills, discover the latest research projects and get to know Trankribus users from around the world. There will also be dinners, coffee breaks and a fantastic mountain hike to enjoy too.



But what makes TUC particularly special is that it is organised by Transkribus users, for Transkribus users. Each year, we invite researchers, archivists, students, and other users to contribute their ideas for workshops, talks and presentations during the conference. All the proposals are reviewed and the users who submit the best ideas then get to present at the TUC.&nbsp;



<h2 class="wp-block-heading">How can I contribute?</h2>



First, you need to think of a topic, one that is useful and/or interesting to other Transkribus users. Perhaps you have a particularly unique way of training models or a good system for collaborating remotely on the same project. Maybe you’ve just embarked on a fascinating research project. It doesn’t matter if you are a complete beginner or a Transkribus expert, what matters is that your topic is interesting and teaches other users something new about Transkribus.



Once you have your topic, then you need to decide on a format.



<h2 class="wp-block-heading">What kinds of formats are there?</h2>



There will be several types of session formats during the conference. Here is a quick overview:



Presentations — These are standard presentations of 20 mins plus 10 mins Q+A. They can be on any aspect of Transkribus, from technical challenges to philosophical discussions.



Scholarship Presentations — These shorter presentations of 10 mins plus 5 mins Q+A give Transkribus scholarship recipients the chance to present their research projects. Five presenters will be chosen and each will receive 10,000 credits as a reward.



Workshops — Share your Transkribus skills with the community. TUC workshops can be on any technical aspect of Transkribus and for any skill level: beginner, intermediate, or advanced. Each workshop lasts approximately 1-2 hours.



Lightning Talks — Want to contribute but don’t have the resources for a full workshop or presentation? In a lightning talk of 5 mins plus 5 mins Q+A, you can present the most important aspects of your research and get instant feedback from other users.



Poster Presentations — Create an A1- or A2-sized poster about your research or project, hang it in the hallway and let other interested users come to you with their questions. It also helps to prepare a quick <a href="https://graduateschool.nd.edu/assets/32665/elevator_pitch_presentation.pdf">elevator pitch</a> too.



<h2 class="wp-block-heading">How do I contribute a topic?</h2>



You should send us your contribution in the form of an academic abstract. If you are a researcher, you probably already have an abstract prepared for your project. If you are not a researcher, the University of Edinburgh has an <a href="http://www.docs.hss.ed.ac.uk/iad/Researchers/Research_staff/Study_Guide_How_to_Write_an_Effective_Abstract_v2.0_.pdf">excellent guide</a>, including examples.



The length of your abstract will depend on which format you wish to contribute. Visit our main <a href="https://readcoop.eu/tuc22/">TUC page</a> to find out more.



<h2 class="wp-block-heading">I’m just a PhD student, can I still contribute something?</h2>



Yes, of course! Every user is welcome to contribute to the TUC, no matter if you’re a prize-winning professor, a professional archivist, an amateur historian or a student. What is important is that your topic is interesting and beneficial to others.



<h2 class="wp-block-heading">How can I submit a contribution?</h2>



All contributions should be submitted online via the main <a href="https://readcoop.eu/tuc22/#call">TUC22 page</a>.



<h2 class="wp-block-heading">When do you need my contribution by?</h2>



The good news is that we have just extended the deadline for contributions until 15 June. We will then read through all the abstracts and let you know if your contribution has been accepted by the 15th of July.



Still have questions? Then please feel free to contact us at <a href="mailto:info@readcoop.eu">info@readcoop.eu</a>.



We look forward to reading your contributions!

Want to contribute to TUC22? The deadline has just been extended!

The new version of the nopaque software programme has just been released, now with a completely new Transkribus HTR pipeline.



<h2 class="wp-block-heading" id="h-what-is-nopaque">What is nopaque?</h2>



nopaque is a custom-built web application developed by a small, interdisciplinary team at the University of Bielefeld, called Data Infrastructure and Digital Humanities (”INF”). The application makes it easy for researchers to get more out of their images and texts without having to worry about the technical side of things.



<h2 class="wp-block-heading" id="h-now-integrated-with-the-power-of-the-transkribus-htr">Now integrated with the power of the Transkribus HTR</h2>



With Transkribus&#8217; help, nopaque has reached one of their long term goals: to integrate an HTR service into nopaque. The Transkribus HTR pipeline service is implemented as a kind of proxied service, which means that the work is split between nopaque and Transkribus. nopaque takes care of preprocessing, storage, and post-processing, while Transkribus handles the recognition and extraction of handwritten and printed texts from scanned images.



If you are interested, try out nopaque <a href="https://nopaque.uni-bielefeld.de/">here</a>.

New version of nopaque released: now including a Transkribus HTR pipeline

After the successful introduction of <a href="https://transkribus.eu/lite">Transkribus Lite 2.0</a>, we have now integrated your valuable feedback into the 2.1 release of Transkribus Lite.



<h2 class="wp-block-heading" id="h-new-view-option-for-the-text-editor">New view option for the text editor</h2>



We have added the option to adjust the text-editor layout to your needs. Now you can move the toolbar to the left side of your screen with one click. This helps you to tidy up the user interface when you want to do focused work.



<figure class="wp-block-video"><video autoplay loop muted src="https://s3.eu-central-1.amazonaws.com/readcoop.cis.public-assets.prod/WP/uploads/2022/03/lite-2-1.mp4"></video></figure>



<h2 class="wp-block-heading">Transkribus will remember your configuration </h2>



Once you have everything set up to your needs in the text editor, Transkribus Lite will now store your settings. When you log back into Transkribus Lite, the layout of the editor, the position of the toolbar, and the selected font-size will be right how you left them after your last session of hard work, wrangling those ancient handwritten documents. 



<h2 class="wp-block-heading">Manage all your text recogntion models in Transkribus Lite</h2>



With the new model manager, you can curate all your models directly in your browser. The model manager lets you change the name, description, languages, and centuries of your models.



<figure class="wp-block-image size-large"><img decoding="async" width="1024" height="597" src="https://s3.eu-central-1.amazonaws.com/readcoop.cis.public-assets.prod/WP/uploads/2022/03/image-1-1024x597.png" alt="" class="wp-image-27989" srcset="https://s3.eu-central-1.amazonaws.com/readcoop.cis.public-assets.prod/WP/uploads/2022/03/image-1-1024x597.png 1024w, https://s3.eu-central-1.amazonaws.com/readcoop.cis.public-assets.prod/WP/uploads/2022/03/image-1-300x175.png 300w, https://s3.eu-central-1.amazonaws.com/readcoop.cis.public-assets.prod/WP/uploads/2022/03/image-1-768x448.png 768w, https://s3.eu-central-1.amazonaws.com/readcoop.cis.public-assets.prod/WP/uploads/2022/03/image-1-1536x896.png 1536w, https://s3.eu-central-1.amazonaws.com/readcoop.cis.public-assets.prod/WP/uploads/2022/03/image-1-2048x1195.png 2048w, https://s3.eu-central-1.amazonaws.com/readcoop.cis.public-assets.prod/WP/uploads/2022/03/image-1-600x350.png 600w" sizes="(max-width: 1024px) 100vw, 1024px" /><figcaption>The new Models Manager</figcaption></figure>



<h2 class="wp-block-heading">Batch delete multiple pages </h2>



In case you want to delete more than one page from a document, you can simply select all the pages you need to remove by ticking them, and delete all of them in one go.

Transkribus Lite 2.1, improved text editor &#038; new models manager

With the start of 2022 we hit another milestone: More than 100 institutions and private persons are now <a href="https://readcoop.eu/members/">members</a> of READ-COOP. This is fully in line with with why we are a co-operative:



<blockquote class="wp-block-quote">Achieve together what one cannot do alone.</blockquote>



A co-operative society, like we are, combines the best of two worlds. We can be active in the market like a company and earn profits, and at the same time we are democratically organised like an association and headed by a board of directors elected by our members.



All our profits are re-invested towards our common goal: revolutionising handwritten text recognition and empowering users. There is no distribution of monetary profits or short-sighted business practices.



We are thrilled, that so many share this vision and want to actively be part of the organisation behind Transkribus. If you want to know how you can join, read more <a href="https://readcoop.eu/join/">here</a>.



Next stop: 200 members 🤞



Driving the historical documents revolution, together 🤝💪

We are 100+!

We are happy to introduce a new PyLaia print model (<a href="https://readcoop.eu/model/transkribus-print-multi-language-dutch-german-english-finnish-french-swedish-etc/">Transkribus print 0.3</a>). You may already be familiar with our HTR+ print model, which in addition to common Antiqua and Fraktur typefaces can also decipher typewritten text, modern computer printouts, and even various unusual ‘decorative fonts’ in several languages. A similar model is now also available for PyLaia. 



We have compared the two models and the results of the PyLaia model seem to match and in some cases surpass those of the HTR+ model. On one of our test sets for example, the new PyLaia model was 30% faster while having a CER of 1.28% compared to 1.64% of the HTR+ model. We have observed before that PyLaia seems to be doing very well on large and diverse train sets such as this. Below you can see some results of the new model, but the best way to see what the model is capable of is to just try it out.&nbsp;



Also, for those of you who have to keep an eye on project budgets: HTR processing with Pylaia models uses slightly fewer credits than with HTR+ models.



<div class="wp-block-image"><figure class="aligncenter size-full"><img decoding="async" width="605" height="183" src="https://s3.eu-central-1.amazonaws.com/readcoop.cis.public-assets.prod/WP/uploads/2021/09/PyLaia-model.jpg" alt="" class="wp-image-26570" srcset="https://s3.eu-central-1.amazonaws.com/readcoop.cis.public-assets.prod/WP/uploads/2021/09/PyLaia-model.jpg 605w, https://s3.eu-central-1.amazonaws.com/readcoop.cis.public-assets.prod/WP/uploads/2021/09/PyLaia-model-300x91.jpg 300w, https://s3.eu-central-1.amazonaws.com/readcoop.cis.public-assets.prod/WP/uploads/2021/09/PyLaia-model-18x5.jpg 18w, https://s3.eu-central-1.amazonaws.com/readcoop.cis.public-assets.prod/WP/uploads/2021/09/PyLaia-model-600x181.jpg 600w, https://s3.eu-central-1.amazonaws.com/readcoop.cis.public-assets.prod/WP/uploads/2021/09/PyLaia-model-190x57.jpg 190w, https://s3.eu-central-1.amazonaws.com/readcoop.cis.public-assets.prod/WP/uploads/2021/09/PyLaia-model-298x90.jpg 298w" sizes="(max-width: 605px) 100vw, 605px" /></figure></div>



<div class="wp-block-image"><figure class="aligncenter size-full"><img decoding="async" width="605" height="322" src="https://s3.eu-central-1.amazonaws.com/readcoop.cis.public-assets.prod/WP/uploads/2021/09/PyLaia-model_2.jpg" alt="" class="wp-image-26571" srcset="https://s3.eu-central-1.amazonaws.com/readcoop.cis.public-assets.prod/WP/uploads/2021/09/PyLaia-model_2.jpg 605w, https://s3.eu-central-1.amazonaws.com/readcoop.cis.public-assets.prod/WP/uploads/2021/09/PyLaia-model_2-300x160.jpg 300w, https://s3.eu-central-1.amazonaws.com/readcoop.cis.public-assets.prod/WP/uploads/2021/09/PyLaia-model_2-18x10.jpg 18w, https://s3.eu-central-1.amazonaws.com/readcoop.cis.public-assets.prod/WP/uploads/2021/09/PyLaia-model_2-600x319.jpg 600w, https://s3.eu-central-1.amazonaws.com/readcoop.cis.public-assets.prod/WP/uploads/2021/09/PyLaia-model_2-113x60.jpg 113w, https://s3.eu-central-1.amazonaws.com/readcoop.cis.public-assets.prod/WP/uploads/2021/09/PyLaia-model_2-169x90.jpg 169w" sizes="(max-width: 605px) 100vw, 605px" /></figure></div>

Check out the new PyLaia model for printed text

On the Maritime Portal there are several reports on the English archive collection of Prize Papers, documents that British privateers captured from enemy ships in the period 1652-1815. The portal also contains information on the website of the Dutch pricing papers, which Huygens ING (KNAW) started two years ago. 72,000 scans of mainly Dutch documents from the 17th to early 19th centuries are freely available at <a href="https://prizepapers.huygens.knaw.nl/">https://prizepapers.huygens.knaw.nl/</a>. The first aim of the Huygens ING-Project is to digitize a great variety of Dutch documents in order to make them freely available in a structured research environment. This was made possible by grants from Bureau Metamorfoze and the Samenwerkende Maritieme Fondsen. The 87 inventory numbers, from which the so diverse, unique, non-loanable material comes, have been provided by Huygens ING with some metadata such as ship and captain names and a periodization, mostly the year in which the (Dutch) ship was captured. This makes it easy to search for inventory numbers, capture dates and names of ships and captains. The requested documents will then appear in the results screen. The website was presented during the first &#8220;Vlootschouw&#8221; symposium in the Scheepvaartmuseum in Amsterdam (January 2020) to a wide audience. In the course of this, the second aim was discussed, namely to make the entire content of the documents searchable and suitable for interdisciplinary analytical research. In order to do so, an automatic transcription (OCR) was used, which made it possible to recognize printed texts. To fully read handwritten documents from the 17th to the 19th centuries, however, another method is needed. This is where Transkribus comes into play. With the HTR software, 100 scans of documents of different periods and languages are used as training material to create a Ground Truth; this happens manually. After the Ground Truth status is achieved, 900 further scans will function as a validation set. The aim is to create a new model with Transkribus which can read the Dutch Prize Papers. Read more here:<a href="https://maritiemportal.nl/blog-dutch-prize-papers-huygens-ing-knaw/#more-19325"> https://maritiemportal.nl/blog-dutch-prize-papers-huygens-ing-knaw/#more-19325&nbsp;</a>

Exploring the Dutch Prize Papers

<h4 class="wp-block-heading" id="h-transkribus-is-happy-to-introduce-new-public-models-for-spanish-and-french">Transkribus is happy to introduce new public models for Spanish and French. </h4>



Álvaro Cuéllar of the ETSO “Estilometría aplicada al Teatro del Siglo de Oro” project&nbsp; trained two models for theater manuscripts and theatre prints from the 17th and 18th century with the aim to modernize the spelling to the current rules. “<a href="https://readcoop.eu/model/spanish-golden-age-theatre-prints-1-0/">The Spanish Golden Age Theatre Manuscripts (Spelling Modernization) 1.0</a>” was trained with theatre manuscripts from the 17th century and the second model on theater prints “<a href="https://readcoop.eu/model/spanish-golden-age-theatre-prints-spelling-modernization-1-0/">Spanish Golden Age Theatre Prints (Spelling Modernization) 1.0</a>” was trained from prints of the 17th and 18th century. 



<div class="wp-block-image"><figure class="aligncenter size-full"><img decoding="async" width="657" height="578" src="https://s3.eu-central-1.amazonaws.com/readcoop.cis.public-assets.prod/WP/uploads/2021/07/french.png" alt="" class="wp-image-26114" srcset="https://s3.eu-central-1.amazonaws.com/readcoop.cis.public-assets.prod/WP/uploads/2021/07/french.png 657w, https://s3.eu-central-1.amazonaws.com/readcoop.cis.public-assets.prod/WP/uploads/2021/07/french-300x264.png 300w, https://s3.eu-central-1.amazonaws.com/readcoop.cis.public-assets.prod/WP/uploads/2021/07/french-14x12.png 14w, https://s3.eu-central-1.amazonaws.com/readcoop.cis.public-assets.prod/WP/uploads/2021/07/french-600x528.png 600w, https://s3.eu-central-1.amazonaws.com/readcoop.cis.public-assets.prod/WP/uploads/2021/07/french-68x60.png 68w, https://s3.eu-central-1.amazonaws.com/readcoop.cis.public-assets.prod/WP/uploads/2021/07/french-102x90.png 102w" sizes="(max-width: 657px) 100vw, 657px" /><figcaption>Sample from the &#8220;New France&#8221; Model of Maxime Gohier</figcaption></figure></div>



There are also two new French models trained by Maxime Gohier: The “<a href="https://readcoop.eu/model/new-france-17th-18th-centuries/">New France 17th-18th centuries</a>” model is a combination of French handwritings from the late 17th and early 18th century and is based on more than 1,600 pages of correspondence and registers of colonial administrators in New France. It includes the main types of writing of the period. The second model “<a href="https://readcoop.eu/model/ordonnances-des-intendants/">Ordonnances des Intendants</a>” was developed from the &#8220;cahiers&#8221; (notebooks) of the ordinances of the intendants of New France (kept at Bibliothèque et Archives nationales du Québec). Written between 1705 and 1760, these notebooks present a good writing regularity and a standardized legal vocabulary.

New models out!

A collaboration between the Bentham Project of the University College London and the DEEDS (Documents of Early England Data Set) Project of the University of Toronto uses Transkribus for the transcription of an immense corpus of medieval charters from the 12th to the 15th century. The handwritten Latin of this period is very peculiar and confronted them with two interesting questions:&nbsp;



<ul><li>Could Transkribus be trained consistently to process abbreviated Latin words, which can represent up to half the vocabulary of medieval legal texts, and hence feature in a substantial proportion of the Documents of Early England Data Set (DEEDS) corpus at the University of Toronto?</li><li>Could Transkribus be made consistently to recognise hyphenated words which span multiple lines of text (insofar as they are both in Latin and abbreviated)?</li></ul>



To find answers, the team first decided to create their own dictionary of over a hundred abbreviated Latin words, both in their abbreviated and in their expanded form. This was done with the help of the independent programmer Ismail Prada from Switzerland, who coded abbrevSolver-master, a Python script. The contracted form was represented by compatible special characters that best reflect how they appear in type. These abbreviations were also categorized as prefixes, suffixes, or standalone abbreviations, which would alter how they would be processed by the algorithm. However, the method turned out to be problematic, as several versions of the appropriate tab-separated Excel file containing the abbreviated words and several varieties of special characters had to be created in an attempt to get it to function as intended. The only way to solve this problem was to proceed with the finding-and-replacing of the abbreviated words without the use of the script by manually finding and replacing the words. This meant a very time intensive process and was not viable in the long run. With the help of Prada, however, the script was fixed and even a superior API script was developed, which is directly connected to Transkribus after giving it the collection editor’s username and password and the collection ID. The new script is quicker and simpler to use. After running a basic command, the script communicates with Transkribus and uses its find-and-replace algorithm on each subcollection, replacing each term it finds from the abbreviation dictionary with its shorter equivalent and tagging them as abbreviated. At this stage of the project, five new HTR models were created. Over the course of this project the WER and CER both declined in a very promising way and the models which were generated after the new script was created, are extremely good. Additionally, the research team used material from Oxford University and Christ Church to further expand the ground truth and was able to create two more models, which improved the testing results of the DEEDS corpus. On the way to the new models, some obstacles, such as poor image quality and the brevity of the images, made the development even more difficult. However, the #7 model is now available freely for everyone. More than 140 000 words have been trained and the CER on the Validation Set is 0.8% For more details about the project and the developed models visit their website: <a href="https://blogs.ucl.ac.uk/transcribe-bentham/2021/04/20/ucl-university-of-toronto-transkribus-htr-and-medieval-latin-abbreviations/">https://blogs.ucl.ac.uk/transcribe-bentham/2021/04/20/ucl-university-of-toronto-transkribus-htr-and-medieval-latin-abbreviations/</a>

Mastering Latin Abbreviations and Hyphenations &#8211; The Bentham and DEEDS Projects

Help us to understand the institutional use of HTR: Joe Nockels is working on this project for his PHD (a Scottish Graduate School of Arts and Humanities (SGSAH) funded PhD). He aims to identify how to incorporate the use of Handwritten Text Recognition (HTR) at scale, working in collaboration with the National Library of Scotland (NLS).



The recent uptake of Handwritten Text Recognition (HTR) in libraries, archives, and other memory institutions has the potential to change access to the written past, at scale. What scale of work using Transkribus (one HTR software) is occurring at institutions? How are the products produced from this work being used? To what degree do institutions feel that the use of Transkribus can be extended? We invite you to take part in a survey that will help us document the changes the use of HTR is making to the information environment, while also thinking of how best to develop Transkribus in a way which meets the need of heritage institutions. The survey should take 10 minutes to complete, and we are particularly interested in responses from those working with Transkribus in institutional settings. The survey can be found <a href="https://edinburgh.onlinesurveys.ac.uk/understanding-institutional-htr-use">here</a>.

Understanding Institutional use of HTR: Your experience with Transkribus

Imagine you are a student who wants to dive into the personal story of one of the few famous child authors in history; or who wants to discover what made the authors of the Spanish Golden Age of theatre tick; or you are a teacher who wants to help the future generation in the digital humanities to achieve such things. There is a programme that may be interesting to those inclined this way. We are proud to be able to support future scholars in their research of historical documents and to encourage them in their projects. Through the Transkribus Scholarship Programme we enable students and also teachers, to use Neural Handwritten Text Recognition in their projects without worrying about funding too much. Transkribus Scholarships are granted based on dissertation and thesis projects, or courses taught and workshops held. Students may apply for HTR credits with a specific project such as a bachelor, master or PhD thesis. In addition, we also think that it is important to support teachers who want to foster the next generation of digital scholars. Therefore, we also support teachers with Scholarships.



<figure class="wp-block-image size-large"><img decoding="async" width="1024" height="576" src="https://s3.eu-central-1.amazonaws.com/readcoop.cis.public-assets.prod/WP/uploads/2021/05/scholarships-1024x576.jpg" alt="" class="wp-image-23692" srcset="https://s3.eu-central-1.amazonaws.com/readcoop.cis.public-assets.prod/WP/uploads/2021/05/scholarships-1024x576.jpg 1024w, https://s3.eu-central-1.amazonaws.com/readcoop.cis.public-assets.prod/WP/uploads/2021/05/scholarships-300x169.jpg 300w, https://s3.eu-central-1.amazonaws.com/readcoop.cis.public-assets.prod/WP/uploads/2021/05/scholarships-768x432.jpg 768w, https://s3.eu-central-1.amazonaws.com/readcoop.cis.public-assets.prod/WP/uploads/2021/05/scholarships-1536x864.jpg 1536w, https://s3.eu-central-1.amazonaws.com/readcoop.cis.public-assets.prod/WP/uploads/2021/05/scholarships-16x9.jpg 16w, https://s3.eu-central-1.amazonaws.com/readcoop.cis.public-assets.prod/WP/uploads/2021/05/scholarships-600x338.jpg 600w, https://s3.eu-central-1.amazonaws.com/readcoop.cis.public-assets.prod/WP/uploads/2021/05/scholarships.jpg 1920w" sizes="(max-width: 1024px) 100vw, 1024px" /><figcaption>Projects that we have supported with our programme.</figcaption></figure>



So far, scholarships were granted to more than 70 applicants in 26 different countries. Many projects in Germany, the Netherlands, Switzerland, the United Kingdom, and other European countries were supported. However, not only European applicants applied for the Scholarship programme, but also students and teachers from the US, China, Japan, Brasil, Singapore, and Israel. With this programme, we enable students to work on smaller projects which are nevertheless very important for the field, and this has made many interesting studies possible.&nbsp;



Joe Nockels, a PhD Student at the University of Edinburgh, working on a project at the National Library of Scotland, applied with his project concerning the diaries of the acclaimed child author Marjory Fleming (1803 – 1811). He used the granted credits to train an HTR+ model entitled “Early 19th Century (Child)”. With an error rate of 1.85 % on the training set and 11.26 % on the validation set, the accuracy of the model is impressive despite the limited amount of diary pages which could be used to train the model. 



<figure class="wp-block-image size-large"><img decoding="async" width="1024" height="601" src="https://s3.eu-central-1.amazonaws.com/readcoop.cis.public-assets.prod/WP/uploads/2021/03/Majory_Fleming-1024x601.jpg" alt="" class="wp-image-21670" srcset="https://s3.eu-central-1.amazonaws.com/readcoop.cis.public-assets.prod/WP/uploads/2021/03/Majory_Fleming-1024x601.jpg 1024w, https://s3.eu-central-1.amazonaws.com/readcoop.cis.public-assets.prod/WP/uploads/2021/03/Majory_Fleming-300x176.jpg 300w, https://s3.eu-central-1.amazonaws.com/readcoop.cis.public-assets.prod/WP/uploads/2021/03/Majory_Fleming-768x451.jpg 768w, https://s3.eu-central-1.amazonaws.com/readcoop.cis.public-assets.prod/WP/uploads/2021/03/Majory_Fleming-16x9.jpg 16w, https://s3.eu-central-1.amazonaws.com/readcoop.cis.public-assets.prod/WP/uploads/2021/03/Majory_Fleming-600x352.jpg 600w, https://s3.eu-central-1.amazonaws.com/readcoop.cis.public-assets.prod/WP/uploads/2021/03/Majory_Fleming.jpg 1422w" sizes="(max-width: 1024px) 100vw, 1024px" /><figcaption>read&amp;search platform of Marjory Fleming&#8217;s writings. </figcaption></figure>



The exported transcription is now easily accessible online via the National Library of Scotland&#8217;s Data Foundry for reuse, research, and further experimentation, and was piloted on the <a href="https://transkribus.eu/r/marjory-fleming/#/">read&amp;search platform</a>. Creating a searchable online version of Fleming’s writings means an important contribution to the historical domain, as with her works the often-neglected voice of children in history takes centre stage and they provide a key source for situation histories of Scotland. 



Another interesting project was carried out by Álvaro Cuéllar (University of Kentucky / Universidad de Valladolid). He is using Transkribus for the automatic transcription and modernization of Spanish Golden Age theatre prints and manuscripts. For the analysis of the texts he uses stylometry and thereby tries to shed light on authorship issues, as it is presented on the <a href="http://etso.es/">etso.es</a> portal. Álvaro is using Transkribus as a fundamental part of his ongoing thesis project about the application of Digital Humanities to Spanish Golden Age theatre.&nbsp;



Those are only two of many highly interesting projects that Transkribus has been able to support. If you yourself are interested or you want to find out more about Transkribus Scholarships, check out the details <a href="https://readcoop.eu/transkribus/scholarship/">here</a>.&nbsp;

Supporting Future Scholars: The Transkribus Scholarship Programme

The municipal archive of the city of Bautzen holds an extensive record of official administration from the 13th century to the present day. Council minutes are part of the main archive collections of the city archive and were almost completely digitized in 2019/2020. They can be accessed online via the <a href="https://www.archivverbund-bautzen.findbuch.net/php/main.php#3632313030">Findbuch.net-platform</a> . However, the statistics of the page views show that they are not too much in use yet. This is certainly also due to the fact, that they are partly difficult to read. With the help of Transkribus and automated text recognition the accessibility of these documents is now to be improved. Furthermore, our web-based read&amp;search-application will be used to make the documents searchable for special words and phrases and this way even more useful for scholars and the interested public.



While the first documents date back to the 13th century, the archive collections also contain extensive series of official records (from the 16th or 17th century). In addition to the court and tax books, the council minutes are particularly noteworthy among the series of official records. In the minutes, which are complete between 1623 and 1832, all decisions of the city council are compactly documented. In particular, they deal with the composition of the council, political decisions, building measures, health and social welfare measures and educational, sales and inheritance matters. &nbsp;



With Transkribus and read&amp;search, we are hoping to contribute to improving the accessibility of the material significantly, as those are vital sources for the history of the city of Bautzen and the surrounding region. This is the link to the website of the Bautzen archive where you can find out more: <a href="https://www.archivverbund-bautzen.de/">https://www.archivverbund-bautzen.de/</a>



<h4 class="wp-block-heading" id="h-update">Update</h4>



As of 2023, it has become possible to access all 256 volumes of the minutes of the Bautzen City Council online. View and search the documents yourself in the online <a href="https://transkribus.eu/r/bautzen-ratsprotokolle/#/">city archive of Bautzen</a> or visit the <a href="https://www.bautzen.de/presse/2023-68">city´s website</a> for more information on the project. 


<div class="wp-block-image">
<figure class="aligncenter size-large"><img decoding="async" width="592" height="374" src="https://s3.eu-central-1.amazonaws.com/readcoop.cis.public-assets.prod/WP/uploads/2021/04/Bautzen-Archiv-2.jpg" alt="" class="wp-image-21822" srcset="https://s3.eu-central-1.amazonaws.com/readcoop.cis.public-assets.prod/WP/uploads/2021/04/Bautzen-Archiv-2.jpg 592w, https://s3.eu-central-1.amazonaws.com/readcoop.cis.public-assets.prod/WP/uploads/2021/04/Bautzen-Archiv-2-300x190.jpg 300w, https://s3.eu-central-1.amazonaws.com/readcoop.cis.public-assets.prod/WP/uploads/2021/04/Bautzen-Archiv-2-16x10.jpg 16w, https://s3.eu-central-1.amazonaws.com/readcoop.cis.public-assets.prod/WP/uploads/2021/04/Bautzen-Archiv-2-95x60.jpg 95w, https://s3.eu-central-1.amazonaws.com/readcoop.cis.public-assets.prod/WP/uploads/2021/04/Bautzen-Archiv-2-142x90.jpg 142w" sizes="(max-width: 592px) 100vw, 592px" /></figure></div>


Source: <a href="https://www.archivverbund-bautzen.de/">https://www.archivverbund-bautzen.de/</a>

Transkribus at the municipal archive of the city of Bautzen

+ Transkribus webinar for the University of Denver Libraries

Already got to know one of our biggest models in Transkribus? That is the German_Kurrent_17th-18th model by the University of Greifswald. Different kinds of texts have been part of the training to do justice to its nickname “The Kurrent Jack of all Trades”. Most of the trained manuscripts are part of the council minutes of the University of Greifswald, the assessor votes of the Wismar High Court, the minutes of the Pomeranian government of Stralsund, the responsa of the Greifswald Law Faculty and other archival holdings such as private letter collections. The training data primarily consists of documents from the 17th-18th century and a couple of pages from the 19th century. The model has a training set of more than 1&nbsp;840&nbsp;000 words (honestly!) and a Character Error Rate on the Validation Set of 5.5%. It is a good choice to get a reasonable transcription for any Kurrent-script you have or if you need a stable base model for 17th-19th century German writings.



Many thanks to the University of Greifswald for this powerful giant!



<figure class="wp-block-image size-large"><img decoding="async" width="756" height="517" src="https://s3.eu-central-1.amazonaws.com/readcoop.cis.public-assets.prod/WP/uploads/2021/02/hanse.jpg" alt="" class="wp-image-19060" srcset="https://s3.eu-central-1.amazonaws.com/readcoop.cis.public-assets.prod/WP/uploads/2021/02/hanse.jpg 756w, https://s3.eu-central-1.amazonaws.com/readcoop.cis.public-assets.prod/WP/uploads/2021/02/hanse-300x205.jpg 300w, https://s3.eu-central-1.amazonaws.com/readcoop.cis.public-assets.prod/WP/uploads/2021/02/hanse-600x410.jpg 600w, https://s3.eu-central-1.amazonaws.com/readcoop.cis.public-assets.prod/WP/uploads/2021/02/hanse-88x60.jpg 88w, https://s3.eu-central-1.amazonaws.com/readcoop.cis.public-assets.prod/WP/uploads/2021/02/hanse-132x90.jpg 132w" sizes="(max-width: 756px) 100vw, 756px" /></figure>

+ &#8220;German_Kurrent_17th-18th&#8221; – The Kurrent Jack of all Trades

Guess who is familiar with a thousand different writers, speaks German, Lower German and Latin and has a favour for 17th century manuscripts? That is the “Acta_17”-model created by Dirk Alvermann, Elisabeth Heigl and Anna Brandt of the University of Greifswald. Thanks to them the model is publicly available in Transkribus &#8211; feel free to try it out if it fits to your material! The training data is based on legal texts and court writings from the Responsa of the Greifswald Law Faculty and can cope with simple German and Latin abbreviations. There is an HTR+ as well as a PyLaia-version of the model and in both cases nearly 600&nbsp;000 words have been trained. The Character Error Rate is about 6%.



In this regard we would like to thank all users, who have shared their models with the Transkribus crowd for their great work! An overview of all public models you can find <a href="https://readcoop.eu/transkribus/howto/public-models-in-transkribus/">here</a> or directly in Transkribus (via “Tools”-tab -&gt; “Models”).



<figure class="wp-block-image size-large is-resized"><a href="https://s3.eu-central-1.amazonaws.com/readcoop.cis.public-assets.prod/WP/uploads/2021/01/Greifswald-Acta-Modell.png"><img decoding="async" src="https://s3.eu-central-1.amazonaws.com/readcoop.cis.public-assets.prod/WP/uploads/2021/01/Greifswald-Acta-Modell-1024x726.png" alt="" class="wp-image-18582" width="684" height="484" srcset="https://s3.eu-central-1.amazonaws.com/readcoop.cis.public-assets.prod/WP/uploads/2021/01/Greifswald-Acta-Modell-1024x726.png 1024w, https://s3.eu-central-1.amazonaws.com/readcoop.cis.public-assets.prod/WP/uploads/2021/01/Greifswald-Acta-Modell-600x425.png 600w, https://s3.eu-central-1.amazonaws.com/readcoop.cis.public-assets.prod/WP/uploads/2021/01/Greifswald-Acta-Modell-300x213.png 300w, https://s3.eu-central-1.amazonaws.com/readcoop.cis.public-assets.prod/WP/uploads/2021/01/Greifswald-Acta-Modell-768x544.png 768w, https://s3.eu-central-1.amazonaws.com/readcoop.cis.public-assets.prod/WP/uploads/2021/01/Greifswald-Acta-Modell-85x60.png 85w, https://s3.eu-central-1.amazonaws.com/readcoop.cis.public-assets.prod/WP/uploads/2021/01/Greifswald-Acta-Modell-127x90.png 127w, https://s3.eu-central-1.amazonaws.com/readcoop.cis.public-assets.prod/WP/uploads/2021/01/Greifswald-Acta-Modell.png 1259w" sizes="(max-width: 684px) 100vw, 684px" /></a></figure>

+ Acta_17 – Greifswald Supercharged

Beyond 2022 is an exciting project – triggered by a sad accident. In 1922 the Public Record Office in Dublin burned down – and thousands of invaluable historical documents got lost. Beyond 2022 – led by the Trinity College Dublin – is working on a virtual reconstruction of the archive – and its documents. Transkribus is on board.
With Transkribus the project team has developed new models for 19th-century handwritings and achieved nice results. The automated transcriptions shall support the publishing of manuscript collections of the Irish Record Commissioners (1810-1830) held at the National Archives of Ireland (NAI). This is especially important knowing that these volumes contain transcripts of lost originals.
In addition, models have been trained on untidy printed and typewritten text and according to the Beyond 2022-team the results are very satisfying. In their project such models are used to recognise the texts in the Dublin Gazette (1750s-1900s). It has been reported that Transkribus copes well with the columnar layout and has been able to reveal “much more incidental gossip than you might expect in an ‘official’ publication.” The homepage of the project and more information you can access via this link: <a href="https://beyond2022.ie/">https://beyond2022.ie/</a>
<img decoding="async" class="wp-image-18346 aligncenter" src="https://s3.eu-central-1.amazonaws.com/readcoop.cis.public-assets.prod/WP/uploads/2021/01/beyond-2022.png" alt="" width="738" height="332" srcset="https://s3.eu-central-1.amazonaws.com/readcoop.cis.public-assets.prod/WP/uploads/2021/01/beyond-2022.png 1411w, https://s3.eu-central-1.amazonaws.com/readcoop.cis.public-assets.prod/WP/uploads/2021/01/beyond-2022-600x270.png 600w, https://s3.eu-central-1.amazonaws.com/readcoop.cis.public-assets.prod/WP/uploads/2021/01/beyond-2022-300x135.png 300w, https://s3.eu-central-1.amazonaws.com/readcoop.cis.public-assets.prod/WP/uploads/2021/01/beyond-2022-1024x460.png 1024w, https://s3.eu-central-1.amazonaws.com/readcoop.cis.public-assets.prod/WP/uploads/2021/01/beyond-2022-768x345.png 768w" sizes="(max-width: 738px) 100vw, 738px" />
Image source: https://beyond2022.ie/?page_id=2#collaborations

+ Transkribus in the Beyond 2022-project

The Europäische Hansemuseum Lübeck carried out a project with school pupils where the students worked directly on the original sources of the Hanseatic cities via the Transkribus Webinterface. The history of the Hanseatic cities is part of the curriculum in schools but original sources in teaching are rare. With the Transkribus Webinterface the pupils can browse through the historic material from home and perform tasks on the documents like searching for person’s names and marking them.



We are inspired by this beautiful project: not only will the school students learn about the history of the city they live in, they will also get an idea about how historical scholarship works and where the information in their schoolbooks comes from. Moreover, this is a very practical way to get in touch with digitization in a historical context and to see how writings undecipherable at first sight can be recognised with the help of technical tools. Currently they are working on the document “Aufruf zur Flandernblockade” from 1358, which is the earliest written testament of common actions of Hanseatic merchants and cities under their self-description “Deutsche Hanse”.



In the context of this project an explanatory video for Transkribus has been created, which we don’t want to keep back from you. You can watch it here: <a href="https://youtu.be/m9SYdKyJcZ0?t=7">https://youtu.be/m9SYdKyJcZ0?t=7</a>



More information about the project you can find via this link: &nbsp;<a href="https://www.hansemuseum.eu/educational_offers/hansequellen/">https://www.hansemuseum.eu/educational_offers/hansequellen/</a>



<figure class="wp-block-image size-large is-resized"><img decoding="async" src="https://s3.eu-central-1.amazonaws.com/readcoop.cis.public-assets.prod/WP/uploads/2020/12/grafik.png" alt="" class="wp-image-17983" width="723" height="440" srcset="https://s3.eu-central-1.amazonaws.com/readcoop.cis.public-assets.prod/WP/uploads/2020/12/grafik.png 605w, https://s3.eu-central-1.amazonaws.com/readcoop.cis.public-assets.prod/WP/uploads/2020/12/grafik-600x365.png 600w, https://s3.eu-central-1.amazonaws.com/readcoop.cis.public-assets.prod/WP/uploads/2020/12/grafik-300x182.png 300w" sizes="(max-width: 723px) 100vw, 723px" /></figure>



©Europäisches Hansemuseum

+ Distance learning with Transkribus

The Hanse-cities, a trade-group from the 13th to the 17th century, have left huge amounts of highly valueable historical documents to be discovered. Until today only a small portion of these documents has been researched by historians.



Thanks to the “Hanse.Quellen.Lesen!”-project many more documents – and especially those from the 16th and 17th century are getting more attention now. Due to the sheer amount , it would not be possible to handle the workload without the help of volunteers &nbsp;and automated transcription. Therefore, the Europäische Hansemuseum Lübeck and the Forschungsstelle für die Geschichte der Hanse und des Ostseeraums (FGHO) started a citizen science project in cooperation with the Archiv der Hansestadt Lübeck (AHL) to work on the transcription with the help of volunteers and Transkribus.



The “first round” of the project has been a spontaneous initiative during the first Corona lockdown in spring 2020. The objective was to try out &nbsp;if a citizen science project would be suitable to work on the scripts. Transkribus has proved to be a suitable tool to provide the technical background for digitization and transcription. Due to the great success the project has moved on to the “second round” by now, called: “Hanse.Quellen.Lesen! 2.0”. FGHO will make its documents available via the read&amp;search application of Transkribus.



Last but not lease it is a great pleasure for us to welcome the Hansemuseum also as a READ-COOP-member and are looking forward to a fruitful collaboration. &nbsp;Events like a Citizen Science Day, a summer school and university seminars dealing with the topics of the “Hanse-Quellen-Lesen!”-project are already on the agenda of our partners from the Hansemuseum.



Here you can find more information about their inspiring project: <a href="https://fgho.eu/de/hanse-quellen-lesen">https://fgho.eu/de/hanse-quellen-lesen</a>



read&amp;search demo version: <a href="https://transkribus.eu/r/demo-read-search/">https://transkribus.eu/r/demo-read-search/</a>

+ Finding out more about undiscovered Hanse-documents with the help of Transkribus

A digital future for old news



Two years ago, Erika Kuijpers (Amsterdam University) in collaboration with Judith Pollman (Leiden University) started the ambitious “Chronicling Novelty” project with the intention to digitize some three hundred local chronicles from the Dutch period of 1500-1850. Using the ScanTent, they photographed thousands of pages of chronicles in archives all over the Netherlands. These images were then uploaded on the VeleHanden platform, which is working together with Transkribus. With enthusiastic work of many volunteers, a great part of the pages is already transcribed and thus will become available to other researchers and interested individuals.



They have also filmed a beautiful video (featuring the ScanTent quite prominently J), showing the achievements and success of the project so far.



Watch the video here: 



<figure class="wp-block-embed is-type-link is-provider-chronicling-novelty"><div class="wp-block-embed__wrapper">
<a href="https://chroniclingnovelty.com/">Bekijk hier het filmpje over ons&nbsp;project</a>
</div></figure>



If cou can read Dutch, you may want to contribute to the success of this project: <a href="https://velehanden.nl/projecten/bekijk/details/project/vua_train_kronieken_transkribus">https://velehanden.nl/projecten/bekijk/details/project/vua_train_kronieken_transkribus</a>

+ Chronicling Novelty: A great crowd-sourcing project from the Netherlands

Some of you might already have come across the word “PyLaia” in connection with Transkribus. Actually, it is one more recognition engine, which has been developed by UPVLC (Universitat Politècnica de València).



PyLaia basically requires the same workflow as HTR+. First of all, please request the PyLaia training feature via <a href="mailto:info@readcoop.eu">info@readcoop.eu</a>, by dropping us a short email indicating the email address you are registered with in Transkribus. We will then enable your account for it. Afterwards PyLaia is accessible via the same “Train”-button as HTR+. You can also use the Ground Truth you have already trained with HTR+ for PyLaia, this way you can compare the results of the two trainings directly. Of course, you can also use PyLaia in order to train new material. PyLaia models will be stored together with all other models on the left side of the “Model”-interface and you can choose it for recognition in the same way as with the HTR+ models. Also, the results in Character Error Rate are usually similar.



The price for recognition with PyLaia will be significantly lower than with HTR+. The exact prices will be made public on our homepage (<a href="https://readcoop.eu/">https://readcoop.eu/</a>) soon, as we plan to start the payment model of Transkribus with 19th of October 2020. It will be very interesting for us to see how PyLaia will perform on you documents and therefore we would like to very much encourage you to try it out. Thanks, and keep transcribing! ?

+ Try out Transkribus new recognition software PyLaia!

In 2019 the <a href="https://www.kb.nl/en">KB National Library of the Netherlands</a> welcomed the <a href="https://www.kb.nl/en/organisation/research-expertise/researcher-in-residence">Researcher-in-Residence project</a> <a href="https://lab.kb.nl/dataset/entangled-histories-ordinances-low-countries">Entangled Histories Ordinances of the Low Countries</a>. Within this project, Annemieke Romein, Sara Veldoen and Michel de Gruijter studied early modern legislation, regarding volumes of printed texts. Transkribus was used in this project in order to make early modern printed texts (e.g. Dutch Gothic) readable.



The 108 volumes that were used within Entangled Histories contained thousands of rules from the early modern era. As the indexes of the various volumes were created with different standards and keywords, searching through the texts can be challenging if they need to be compared. Hence, Entangled Histories aimed at dissecting the individual texts and consequently categorise them according to a controlled vocabulary. As segmentation of texts is a field that is heavily under development, a couple of tests on what would work had been run through. In order to automatically categorise the legal texts, the Finnish tool <a href="http://www.annif.org">Annif</a> had been used. This tool has various back-ends that enable a range of options for automatic categorisation. As the project used a controlled, hierarchical vocabulary, they had to create a SKOS for this specific project. Though Annif had not been used at hierarchical structures before, they got some excellent results even though they ended up using only 400 texts in the case study.



Such a combination of techniques – recognition, segmentation and categorisation – could be interesting to also to other projects and archives to be able to quickly metadata individual texts. If you want to know more, you can check out these recent publication in the <a href="http://journal.dhbenelux.org/volume-2-digital-humanities-in-society/">DHBenelux Journal</a> (OA) called: <a href="http://journal.dhbenelux.org/journal/issues/002/article-23-romein/article-23-romein.pdf">The Datafication of Early Modern Ordinances</a>.

+ Entangled Histories Ordinances of the Low Countries

We are happy to present a new public model for Italian administrative hands! It was created as a collaborative effort by Jake Dyble (Exeter/Pisa), Antonio Iodice (Exeter/Genoa), Sara Mansutti (Cork), and Rachel Midura (Virginia Tech). The model “Italian Administrative Hand, 1550-1700” was trained on a variety of Italian-language documents from state archives in Milan, Venice, Florence, Pisa, and Genoa. The training set represents a spectrum of humanistic, italic and cursive hands characteristic of administrative records, employed by secretaries and news writers.



The model has been trained to perform well with a mix of quantitative and qualitative information as well as many common proper nouns for the period, such as locations in Europe and contemporary rulers. Administrative documents often employ common superscript abbreviations, which the accompanying documentation treats in greater detail. The model can also be used with Latin, Spanish and French documents to some extent.



More information can be found at: <a href="https://emdigit.org/tool/2020/07/21/italian-administrative-hands.html">https://emdigit.org/tool/2020/07/21/italian-administrative-hands.html</a>

+ Public model for Italian administrative hands

We would like to make you aware of the Innovation in Archival Research Prize, that is organised by the National Archives of the Netherlands and the North Holland Archives. Within the scope of this research prize, the best demonstration of the value of the availability of large quantities of transcripts from the archives of the VOC and Haarlem notaries will be awarded. It is all about showing what new research or visualization possibilities there are, when archives are making use of automatic handwritten text recognition. The competition is connected to the project Making the Iceberg Visible of the National Archives where large quantities of historical documents are automatically transcribed and converted into machine-readable text with Transkribus technology.



If you are a student and interested in taking part in the competition, submit your research results until April 15th 2021. The award ceremony will take place during a hybrid (partly online) symposium on May 31, 2021, where the winner will present her or his research. More information about the award and the requirements for submission can be looked up here: <a href="https://www.nationaalarchief.nl/handschriftherkenning">https://www.nationaalarchief.nl/handschriftherkenning</a>



PS: Günter Mühlberger will be one of the jury members ?

+ Students up for a research competition?

<h1 class="wp-block-heading"></h1>



The Transkribus community keeps growing, we have passed the 40.000 users mark recently, and so does the number of READ-COOP-members. We currently have more than 60 private persons and institutions, who decided to become a co-owner of READ-COOP SCE. Apart from getting discounts on services of the READ-COOP they also have the chance to actively take part in the decision making and voting at the General Meeting. We are proud that well known institutions such as the British Library, the Trinity College Dublin, the Austrian Academy of Science and the University of Cambridge signed the contract.



Our institutional members mostly have an archival, library or university background, so some of the most recent members are the University of Aberdeen (<a href="https://www.abdn.ac.uk/">https://www.abdn.ac.uk/</a>), the Kiel University Library (<a href="https://www.uni-kiel.de/en/">https://www.uni-kiel.de/en/</a>) and the National Archives Luxembourg (<a href="https://anlux.gouvernement.lu/en.html">https://anlux.gouvernement.lu/en.html</a>). But there are also museums like the Europäische Hansemusem Lübeck (<a href="https://www.hansemuseum.eu/">https://www.hansemuseum.eu/</a>), which are taking part in the READ-COOP. What might be interesting: the Kiel University has published an article about their motivation of becoming a member. The text is in German – if you would like to read it or copy it into some kind of online translator, you can do so via this link: <a href="https://www.kooperation-international.de/aktuelles/nachrichten/detail/info/staerkung-der-digital-humanities-universitaet-kiel-wird-teil-der-internationalen-read-genossenschaft/">https://www.kooperation-international.de/aktuelles/nachrichten/detail/info/staerkung-der-digital-humanities-universitaet-kiel-wird-teil-der-internationalen-read-genossenschaft/</a>



Institutions as well as private persons can become READ-COOP-members and benefit from the advantages connected with that. If you are interested you can get more information here: <a href="https://readcoop.eu/join-us/">https://readcoop.eu/join-us/</a> , if you have more questions, you can always get in touch with us via: <a href="mailto:info@readcoop.eu">info@readcoop.eu</a>. An overview of all the READ-COOP-members you can find on our homepage: <a href="https://readcoop.eu/members/">https://readcoop.eu/members/</a>

+ More than 60 members already support the READ-COOP SCE

Recently an interview on the experience with Transkribus has been publicised by the Gramsci centre for the humanities and the University of San Marino. A model for the prison notebooks of Antonio Gramsci has been created, Massimo Mastrogregori of the Gramsci centre for the humanities has been working on this.



You can check out the whole article (in Italian) via this link: <a href="https://www.gramsciforthehumanities.org/transkribus-analisi-elettronica-testi-quaderni-del-carcere/">https://www.gramsciforthehumanities.org/transkribus-analisi-elettronica-testi-quaderni-del-carcere/</a>



<div class="wp-block-image"><figure class="aligncenter size-large is-resized"><img decoding="async" src="https://s3.eu-central-1.amazonaws.com/readcoop.cis.public-assets.prod/WP/uploads/2020/07/Gramsci_1922.jpg" alt="" class="wp-image-8767" width="316" height="421" srcset="https://s3.eu-central-1.amazonaws.com/readcoop.cis.public-assets.prod/WP/uploads/2020/07/Gramsci_1922.jpg 443w, https://s3.eu-central-1.amazonaws.com/readcoop.cis.public-assets.prod/WP/uploads/2020/07/Gramsci_1922-225x300.jpg 225w" sizes="(max-width: 316px) 100vw, 316px" /><figcaption> Source: <a href="https://en.wikipedia.org/wiki/Prison_Notebooks#/media/File:Gramsci_1922.jpg">https://en.wikipedia.org/wiki/Prison_Notebooks#/media/File:Gramsci_1922.jpg</a> </figcaption></figure></div>

+Transkribus e le frontiere dell’analisi elettronica dei testi

Want to know, what a searching interface based on Transkribus can look like? We have two new demo videos available on that in our Youtube-channel (<a href="https://www.youtube.com/c/TranskribusEU">https://www.youtube.com/c/TranskribusEU</a>) showing two projects, which have been carried out with the goal to make historical documents searchable by special words and phrases.



One video shows the Finnish Court Records of the Finland National Archives as an use case for web-based Keyword-searching: <a href="https://www.youtube.com/watch?v=H9Sp2GuUVmM">https://www.youtube.com/watch?v=H9Sp2GuUVmM</a>



The second video demonstrates the Full-Text-searching-interface of the New Zealand Alpine Journal project: <a href="https://www.youtube.com/watch?v=4u1E3UQbW5M">https://www.youtube.com/watch?v=4u1E3UQbW5M</a>



If you would like to know more about searching in documents with Transkribus you can always get in touch with us via <a href="mailto:info@readcoop.eu">info@readcoop.eu</a>.



<div class="wp-block-image"><figure class="aligncenter size-large is-resized"><img decoding="async" src="https://s3.eu-central-1.amazonaws.com/readcoop.cis.public-assets.prod/WP/uploads/2020/05/NAF-interface.jpg" alt="" class="wp-image-7074" width="419" height="225" srcset="https://s3.eu-central-1.amazonaws.com/readcoop.cis.public-assets.prod/WP/uploads/2020/05/NAF-interface.jpg 801w, https://s3.eu-central-1.amazonaws.com/readcoop.cis.public-assets.prod/WP/uploads/2020/05/NAF-interface-600x322.jpg 600w, https://s3.eu-central-1.amazonaws.com/readcoop.cis.public-assets.prod/WP/uploads/2020/05/NAF-interface-300x161.jpg 300w, https://s3.eu-central-1.amazonaws.com/readcoop.cis.public-assets.prod/WP/uploads/2020/05/NAF-interface-768x412.jpg 768w" sizes="(max-width: 419px) 100vw, 419px" /></figure></div>



<div class="wp-block-image"><figure class="aligncenter size-large is-resized"><img decoding="async" src="https://s3.eu-central-1.amazonaws.com/readcoop.cis.public-assets.prod/WP/uploads/2020/05/Alpine-Journal-Interface.jpg" alt="" class="wp-image-7075" width="416" height="222" srcset="https://s3.eu-central-1.amazonaws.com/readcoop.cis.public-assets.prod/WP/uploads/2020/05/Alpine-Journal-Interface.jpg 788w, https://s3.eu-central-1.amazonaws.com/readcoop.cis.public-assets.prod/WP/uploads/2020/05/Alpine-Journal-Interface-600x321.jpg 600w, https://s3.eu-central-1.amazonaws.com/readcoop.cis.public-assets.prod/WP/uploads/2020/05/Alpine-Journal-Interface-300x160.jpg 300w, https://s3.eu-central-1.amazonaws.com/readcoop.cis.public-assets.prod/WP/uploads/2020/05/Alpine-Journal-Interface-768x410.jpg 768w" sizes="(max-width: 416px) 100vw, 416px" /></figure></div>

+ Videos on Keyword and Full-Text search interfaces

Many models on different languages of various epochs have been trained in Transkribus already. What we are still missing is a general model, which can recognise a wide range of modern handwritings. We have set ourselves the goal to provide such a model to our users in near future: a big and robust model, that can deal with most of modern handwritings.



Thinking about how to create the big amount of training data needed, crowdsourcing came to our mind and the idea for the readme2020 project was born. That’s where you and other Transkribus users come into play: please help us setting up the training data by providing a piece of handwritten text!



If you would like to contribute, the following four steps are everything, which is needed in order to do so:



<ol type="1"><li>Take a blank sheet of paper and write something down by hand, this can be any kind of text and topic. Of course, you can also simply copy a text.</li><li>Download the DocScan-app on your Android mobile phone (unfortunately not available for iPhones) and scan your writing, by following the instructions, shown on the screenshots on this page: <a href="https://readcoop.eu/readme2020/">https://readcoop.eu/readme2020/</a></li><li>Upload the document to Transkribus</li><li>You will get a confirmation-email with the link to our web-interface. Please add the transcription of your text there.</li></ol>



That’s already it.



As soon as there will be enough data, we from the READ-COOP-team will start the training of the modern handwriting model. We are super-curious about the outcome and are looking forward to creating something beneficial for many people together with you.



With that link you can read about all about the project-details: <a href="https://readcoop.eu/readme2020/">https://readcoop.eu/readme2020/</a>



<div class="wp-block-image"><figure class="aligncenter size-large"><img decoding="async" width="735" height="280" src="https://s3.eu-central-1.amazonaws.com/readcoop.cis.public-assets.prod/WP/uploads/2020/05/readme2020.png" alt="" class="wp-image-7035" srcset="https://s3.eu-central-1.amazonaws.com/readcoop.cis.public-assets.prod/WP/uploads/2020/05/readme2020.png 735w, https://s3.eu-central-1.amazonaws.com/readcoop.cis.public-assets.prod/WP/uploads/2020/05/readme2020-600x229.png 600w, https://s3.eu-central-1.amazonaws.com/readcoop.cis.public-assets.prod/WP/uploads/2020/05/readme2020-300x114.png 300w" sizes="(max-width: 735px) 100vw, 735px" /></figure></div>

+ readme2020: crowdsourcing for the recognition of modern handwritings

This project was set up to investigate the local parish records of Devon, some of which date back to the 16th century but the majority of which are 18 and 19th century. These records are largely about the raising of local taxes and then spending the money raised for the upkeep of the church and the welfare of the poor. Providing for the poor has been a legal requirement in the UK since the time of Elizabeth the First and this was managed locally long before any idea of a ‘welfare state’ was born. The records had been digitised (scanned and catalogued) as a result of a previous project and this way several thousand digital images with search facilities have become available.



Many stories were hidden within these records and Roger Claxton had been able to put together a book that picks out some of these stories and includes many illustrations taken from the documents together with the transcriptions. Some details of this book can be found at <a href="https://www.widecombe-in-the-moor.com/welfare/">https://www.widecombe-in-the-moor.com/welfare/</a>



He used Transkribus to create a workflow for the control and transcription of a considerable number of these documents. This workflow and the handwriting models generated can be used again in the future. The writing is fairly typical of the 18th and 19th centuries.



That’s what Roger Claxton says about his work with Transkribus:



“I found Transkribus provided a good structure within which to work both when building up a test set of transcribed documents (by manually entering the transcriptions) and when using Transkribus to transcribe for me once I had a suitable model. It was not perfect but then I would not expect it to be.



The documents were not especially difficult to transcribe but having a framework within which to operate helped enormously. Having the documents available with the tools to view them together with the transcription, which can be worked on, left, and then returned to, helped me keep everything organised.



Once I was happy with the transcription I could copy and paste the transcribed text into my written script.



All in all I have been very pleased to have Transkribus available to use and hope that it can remain available for future use. Thanks to the Transkribus team and to the EU for the support!”

+ Widecombe Parish Devon UK: Investigating Local Welfare Provision

We would like to inform you about an article written by Marco Roling, which has been published in March 2020 and is dedicated to experiments about the HTR-processing of documents with low standard of preservation and legibility. The tests have been carried out on documents with different levels of archival damage in order to check, how Handwritten Text Recognition can deal with that. Moreover, some ideas are presented on how the probability of good HTR-performance can be predicted.



This is the link to the online resource, if you would like to read the whole article: <a href="https://www.researchgate.net/publication/340117708_Does_Handwriting_Text_Recognition_Work_for_Damaged_Archives">https://www.researchgate.net/publication/340117708_Does_Handwriting_Text_Recognition_Work_for_Damaged_Archives</a>

+ Handwritten Text Recognition (HTR) for damaged archives

In the article with the title &#8220;Automated Transcription of Handwritten Text: READ and TRANSKRIBUS (An Experiment with Transcribing Letters of Andrej Kmeť)&#8221;, which was published in October 2019, Dusan Katuscak describes his experience with Transkribus when working on the transcription of the handwritten letters of Andrej Kmet’. Additionally, a brief description of the Transkribus workflow (scanning with the ScanTent, image-upload, segmentation, automatic transcription) is given.



If you are interested to know more, this is the link, where you can have a look at the whole article: <a href="https://www.researchgate.net/publication/336676993">https://www.researchgate.net/publication/336676993</a>

+ Deciphering the letters of Andrej Kmet’ with the help of Transkribus

We from the Transkribus-team keep noticing interest amongst ambitioned users of Transkribus in giving Transkribus-workshops and this way share their knowledge. In order to provide these people with the latest information on Transkribus features, we plan to offer a trainer seminar in Innsbruck. The seminar is currently planned from Tuesday 15th of September until Thursday 17th of September 2020. We will need at least three full days to cover the most important topics. The current curriculum has got 16 modules, so a lot of information for us to run through ?.



Due to the extraordinary Corona-situation we have in many countries at the moment, we might have to postpone the training. Nevertheless, we would like to make you aware, so you can save the date in case you are interested in taking part and we will of course keep you posted on possible changes.



In order to get the most out of the seminar for every participant the maximum number will be limited to 15 people. We already have a couple of people interested in taking part, so please let us know as soon as you can if you would like to join in (via info@readcoop.eu).



We are excited to have a detailed look on Transkribus with you!

+ Train the trainers in Innsbruck

This is what comes out, when two Transkribus power user archives, namely the Amsterdam City Archives and the National Archives of the Netherlands work together: a model with 1&nbsp;384&nbsp;893 words of training data, in this case reading 18th century Dutch. The model is available to all Transkribus users now and can be found under the name: “Dutch Mountains (18th Century)”. It combines the 18th Century models of the two archives (Amsterdam City Archives : 3500+ scans of&nbsp;15 notarial handwritings and National Archives of the Netherlands&nbsp;3500+ scans of&nbsp;VOC handwritings). The Character Error Rate goes down to 5,67%.



When you are using models big as this one for your documents, it makes sense to add the corrisponding language model, or a base model, if you have already trained a model yourself. That’s where you can find the language model setting in Transkribus: “Tools”-tab -> Click “Run” in the “Text Recognition”-section -> “Select HTR-model” -> “Dictionary” (top left) -> “Language model from training data”



Have fun trying out this or one of our other public models! With this link you can have a look at the overview of all the available models: <a href="https://transkribus.eu/wiki/images/d/d6/Public_Models_in_Transkribus.pdf">https://transkribus.eu/wiki/images/d/d6/Public_Models_in_Transkribus.pdf</a>

+ Proudly presenting the Dutch giant

The second Transkribus User Conference took place at the Technical University Vienna on 8-9 November 2018.  Around 120 attendees shared their experiences with Transkribus and learnt about the latest innovations in text recognition technology.
Videos of the conference presentations are <a href="https://www.youtube.com/channel/UC-txVgM31rDTGlBnH-zpPjA">online on YouTube</a>!
<h4>Conference programme</h4>
<ul>
<li><a href="https://read.transkribus.eu/wp-content/uploads/2018/10/tuc-program-2018.pdf">Programme</a></li>
</ul>
<h4>Slides (licensed by <a href="https://creativecommons.org/licenses/by-nc-nd/4.0/" target="_blank" rel="noopener noreferrer">CC BY-NC-ND 4.0</a>)</h4>
Introduction
<ul>
<li>Günter Mühlberger (University of Innsbruck) &#8211; <a href="https://read.transkribus.eu/wp-content/uploads/2018/11/MUHLBERGER-HIGHLIGHTS-.pdf">READ and Transkribus – highlights of 2018 and future outlook</a></li>
<li>Markus Dellinger (Österreichischer Raiffeisenverband) &#8211; <a href="https://read.transkribus.eu/wp-content/uploads/2018/11/DELLINGER-SCE.pdf">The READ-COOP: working together for the future of digital cultural heritage</a></li>
</ul>
Transkribus in the archive &#8211; automated recognition of large historical collections in European archives
<ul>
<li><a href="https://read.transkribus.eu/wp-content/uploads/2018/11/HODEL-ZURICH.pdf">Tobias Hodel (State Archives of Zurich)</a></li>
<li><a href="https://read.transkribus.eu/wp-content/uploads/2018/11/LANG-PASSAU.pdf">Eva Lang (Passau Diocesan Archives)</a></li>
<li><a href="https://read.transkribus.eu/wp-content/uploads/2018/11/KALLIO-HIRVONEN-FINLAND.pdf">Maria Kallio and Lauri Hirvonen (National Archives of Finland)</a></li>
<li><a href="https://read.transkribus.eu/wp-content/uploads/2018/11/PONTE-AMSTERDAM.pdf">Mark Ponte (Amsterdam City Archives)</a> &#8211; crowdsourcing with Transkribus</li>
</ul>
Transkribus in Practice (presentations from Transkribus users)
<a href="https://read.transkribus.eu/wp-content/uploads/2018/11/VAINIO-KURTAKKO-AND-STUBB-SLS.pdf">Maria Vainio-Kurtakko and Elisabeth Stubb (Society of Swedish Literature in Finland)</a>&#8211; papers of Finnish painter Albert Edelfelt (1854-1905)
<a href="https://read.transkribus.eu/wp-content/uploads/2018/11/MASSOT-ENS.pdf">Marie-Laure Massot (CAPHÉS / ANR FFL &#8211; Ecole normale supérieure Paris)</a> &#8211; papers of French philosopher Michel Foucault (1926-1984)
<a href="https://read.transkribus.eu/wp-content/uploads/2018/11/GREENSTREET-MARINE-LIVES.pdf">Colin Greenstreet (Marine Lives &amp; Chronoscopic Education)</a> &#8211; early modern legal papers, signatures, crowdsourcing
<a href="https://read.transkribus.eu/wp-content/uploads/2018/11/BROWN-DAVID-DUBLIN-V2.pdf">David Brown (Trinity College Dublin)</a> &#8211; documents from Irish Public Record Office
<a href="https://read.transkribus.eu/wp-content/uploads/2018/11/ANU-LAHTINEN-HELSINKI.pdf">Anu Lahtinen (University of Helsinki)</a> &#8211; sixteenth-century Swedish administrative documents
<a href="https://read.transkribus.eu/wp-content/uploads/2018/11/SLON-ZAPALA.pdf">Marek Słoń, Mikołaj Słoń and Adam Zapała (Institute of History, Polish Academy of Sciences)</a> &#8211; sixteenth-century tax registers, GIS mapping
<a href="https://read.transkribus.eu/wp-content/uploads/2018/11/DAY-CHRIS-TNA-LONDON.pdf">Chris Day (The National Archives UK)</a>&#8211; early modern wills, language models
<a href="https://read.transkribus.eu/wp-content/uploads/2018/11/BELOBORODOVA-DILLEN-SCHAUABLE.pdf">Olga Beloborodova, Wout Dillen and Joshua Schäuble (University of Antwerp)</a> &#8211; CATCH 2020 project, manuscript genetics, digital scholarly editing, Samuel Beckett (1906 &#8211; 1989)
What’s new in Transkribus? Update on latest features.
<ul>
<li>Gundram Leifert (University of Rostock), <a href="https://read.transkribus.eu/wp-content/uploads/2018/11/LEIFERT-CITLAB.pdf">HTR+</a> (improved form of Handwritten Text Recognition)</li>
<li>Hervé Déjean (Naver Labs Europe), <a href="https://read.transkribus.eu/wp-content/uploads/2018/11/DEJEAN-NAVER.pdf">Table Processing</a></li>
<li>Günter Mühlberger (University of Innsbruck), <a href="https://read.transkribus.eu/wp-content/uploads/2018/11/MUHLBERGER-KWS-SHARING-WEB.pdf">Keyword Spotting, sharing HTR models and Transkribus web</a></li>
</ul>
Digitisation on demand in archives 
<ul>
<li>Florian Kleber, Markus Diem and Fabian Hollaus (TU Wien), <a href="https://read.transkribus.eu/wp-content/uploads/2018/11/KLEBER-DIEM-HOLLAUS-.pdf">DocScan and the ScanTent</a></li>
<li>Dirk Alvermann (University of Greifswald), <a href="https://read.transkribus.eu/wp-content/uploads/2018/11/ALVERMANN-GREIFSWALD.pdf">DocScan at Greifswald &#8211; Idea, practice and mutual benefit of scanning on demand</a></li>
</ul>
<iframe title="DocScan Greifswald" width="800" height="450" src="https://www.youtube.com/embed/BVnKnsWUHOM?feature=oembed" frameborder="0" allow="accelerometer; autoplay; clipboard-write; encrypted-media; gyroscope; picture-in-picture" allowfullscreen></iframe>
Keyword Spotting (in theory)
<ul>
<li>Ioannis Pratikakis (Democritus University of Thrace), <a href="https://read.transkribus.eu/wp-content/uploads/2019/02/PRATIKAKIS-DUTH.pdf">Keyword Spotting: Query by Example</a></li>
<li>Alejandro Héctor Toselli and Enrique Vidal (Universitat Politècnica de València), <a href="https://read.transkribus.eu/wp-content/uploads/2018/11/TOSELLI-ALEJANDRO-UPVLC.pdf">Indexing and Searching of Manuscript Collections</a></li>
</ul>
Keyword Spotting (in practice)
<ul>
<li>Louise Seaward (University College London), <a href="https://read.transkribus.eu/wp-content/uploads/2018/11/SEAWARD-UCL.pdf">Maximising the utility of Jeremy Bentham’s manuscripts</a></li>
</ul>
The Transkribus User Conference was organised as part of the READ project.  This project has received funding from the European Union’s Horizon 2020 research and innovation programme under grant agreement No 674943.
<a href="https://read.transkribus.eu/wp-content/uploads/2018/11/Transkribus_wolpertinger_transparent.png"><img decoding="async" class="aligncenter size-full wp-image-2326" src="https://read.transkribus.eu/wp-content/uploads/2018/11/Transkribus_wolpertinger_transparent.png" alt="" width="555" height="623" /></a>

Transkribus User Conference 2018

The first Transkribus User Conference took place at the <a href="https://www.tuwien.ac.at/en/" target="_blank" rel="noopener noreferrer">Technical University Vienna</a> on 2-3 November 2017.
Around 80 attendees came together to share their experiences with Transkribus, hear about the latest advances in Automated Text Recognition and ask developers how their documents could be automatically processed.
You can find further information in the conference programme, videos and presenters&#8217; slides provided below.
<img decoding="async" class="aligncenter size-large wp-image-1273" src="https://read.transkribus.eu/wp-content/uploads/2017/07/view-tu-the-sky-1024x768.jpg" alt="" width="720" height="540" />
Conference programme
<a href="https://read.transkribus.eu/wp-content/uploads/2017/07/Conference_Programme.pdf">Programme</a>
Videos (licensed by  <a href="https://creativecommons.org/licenses/by-nc-nd/4.0/" target="_blank" rel="noopener noreferrer">CC BY-NC-ND 4.0</a> Ingrid Reinhardt)
Playlist:
<ul>
<li><a href="https://www.youtube.com/watch?v=_7V8iXFzx_E&amp;list=PL7UbQtd4qlhLNv7zYnlyTdnLBsMJw6TOt" target="_blank" rel="noopener noreferrer">Watch all presentations from the conference</a></li>
</ul>
Sessions:
<iframe title="01 Transkribus - the status quo and future plans" width="800" height="450" src="https://www.youtube.com/embed/_7V8iXFzx_E?feature=oembed" frameborder="0" allow="accelerometer; autoplay; clipboard-write; encrypted-media; gyroscope; picture-in-picture" allowfullscreen></iframe>
<iframe title="02 Transkribus in practice. Reports from users." width="800" height="450" src="https://www.youtube.com/embed/qnY3i6d4lHw?feature=oembed" frameborder="0" allow="accelerometer; autoplay; clipboard-write; encrypted-media; gyroscope; picture-in-picture" allowfullscreen></iframe>
<iframe title="03 Keyword searching and indexing" width="800" height="450" src="https://www.youtube.com/embed/APpugG4Yzso?feature=oembed" frameborder="0" allow="accelerometer; autoplay; clipboard-write; encrypted-media; gyroscope; picture-in-picture" allowfullscreen></iframe>
<iframe title="04 More reports from users - Part 1" width="800" height="450" src="https://www.youtube.com/embed/Ih4mGsXT7NU?feature=oembed" frameborder="0" allow="accelerometer; autoplay; clipboard-write; encrypted-media; gyroscope; picture-in-picture" allowfullscreen></iframe>
<iframe title="04 More reports from users - Part 2" width="800" height="450" src="https://www.youtube.com/embed/QE9jeupnK4Y?feature=oembed" frameborder="0" allow="accelerometer; autoplay; clipboard-write; encrypted-media; gyroscope; picture-in-picture" allowfullscreen></iframe>
<iframe title="05 Panel discussion" width="800" height="450" src="https://www.youtube.com/embed/KJ-cp3u2qr4?feature=oembed" frameborder="0" allow="accelerometer; autoplay; clipboard-write; encrypted-media; gyroscope; picture-in-picture" allowfullscreen></iframe>
Presentation slides (licensed by <a href="https://creativecommons.org/licenses/by-nc-nd/4.0/" target="_blank" rel="noopener noreferrer">CC BY-NC-ND 4.0</a>)
Transkribus Users 
<a href="https://read.transkribus.eu/wp-content/uploads/2017/07/Alvermann_Bluegel_Greifswald.pdf">Dirk Alvermann and Bruno Blüggel (Universität Greifswald)</a> &#8211; working on digital search and display of c.18-19 German documents.
<a href="https://read.transkribus.eu/wp-content/uploads/2017/07/Dillen_Beckett_edition.pdf">Wout Dillen (University of Antwerp)</a> &#8211; working on digital edition of Samuel Beckett&#8217;s writings (c.20).
<a href="https://read.transkribus.eu/wp-content/uploads/2017/07/Boenig_Wuerzner_Groundtruth-1.pdf">Matthias Boenig and Kay-Michael Würzner (Berlin-Brandenburg Academy of Sciences and Humanities)</a> &#8211; working on training data to improve recognition of printed text.
<a href="https://read.transkribus.eu/wp-content/uploads/2017/07/Vetter_Austrian_Bible.pdf">Angila Vetter (Universität Augsburg) </a>&#8211; working on critical edition of Austrian Bible.
<a href="https://read.transkribus.eu/wp-content/uploads/2017/07/Thoele_Incunable.pdf" target="_blank" rel="noopener noreferrer">Karen Thöle (University of Göttingen)</a> &#8211; working with a heavily abbreviated medieval incunable.
<a href="https://read.transkribus.eu/wp-content/uploads/2017/07/Valdes_PROLOPE.pdf">Ramón Valdés Gázquez (Universitat Autonoma de Barcelona)</a> &#8211; working on critical edition of Lope de Vega&#8217;s plays.
<a href="https://read.transkribus.eu/wp-content/uploads/2017/07/Cornell_Georgian_Papers.pdf">Deborah Cornell (William &amp; Mary Libraries)</a> &#8211; working on transcription of documents from reign of George III (c.18).
<a href="https://read.transkribus.eu/wp-content/uploads/2017/07/Onasch_Lemke_Barlach.pdf">Karoline Lemke and Paul Onasch (Universität Rostock)</a>&#8211; working on edition of Ernst Barlach&#8217;s (c.19) letters.
<a href="https://read.transkribus.eu/wp-content/uploads/2017/07/Kampkaspar_Wienerisches_Diarium.pdf">Dario Kampkaspar (Austrian Academy of Sciences)</a> &#8211; working on digital Edition of Wienerisches Diarium newspaper (c.18).
<a href="https://read.transkribus.eu/wp-content/uploads/2017/07/Jerkov_Uni_Belgrade_library.pdf">Aleksandar Jerkov (University of Belgrade Library)</a> &#8211; working on automated recognition of Cyrillic text.
<a href="https://read.transkribus.eu/wp-content/uploads/2017/07/Lang_Table_Processing.pdf">Eva Lang (Passau Diocesan Archives)</a> &#8211; working on processing of tables and forms in church registry documents (c.19).
Presentations from READ project partners
<a href="https://read.transkribus.eu/wp-content/uploads/2017/07/Labahn_Keyword_Spotting.pdf">Roger Labahn (Universität Rostock)</a> &#8211; Keyword searching and indexing in large collections of handwritten documents
<a href="https://read.transkribus.eu/wp-content/uploads/2017/07/Toselli_Keyword_Spotting.pdf">Alejandro Toselli (Universitat Politècnica de València)</a> &#8211; Keyword spotting in large scale documents
<a href="https://read.transkribus.eu/wp-content/uploads/2017/07/Hodel_Leifert_t2i_matching.pdf">Gundram Leifert (Universität Rostock) and Tobias Hodel (State Archives of Zurich)</a> &#8211; Text2Image matching: how to use existing images and transcripts to produce Automated Text Recognition
<a href="https://read.transkribus.eu/wp-content/uploads/2017/07/Kleber_Competitions.pdf">Florian Kleber (TU Wien)</a> &#8211; Progress through competition: linking humanities data with computer science research
The Transkribus User Conference was organised as part of the READ project.  This project has received funding from the European Union’s Horizon 2020 research and innovation programme under grant agreement No 674943.

Transkribus User Conference 2017

Our 3rd Transkribus User Conference will take place at the University of Innsbruck on 6-7 February 2020. Information will be shared about our recently founded READ-COOP SCE, latest technical developments and how Transkribus is used by several projects around the world. Also there will be ten workshops to complete the program. And finally there is the chance to go skiing and sledging with Transkribus team members on Saturday 8th February.



Venue



The conference will take place at the Faculty of Catholic Theology of the University of Innsbruck, which is in the city centre of Innsbruck.&nbsp;The address of the conference building is:&nbsp;Karl-Rahner-Platz 3, 6020 Innsbruck.



Registration



Unfortunately registration is not possible any more &#8211; we are fully booked.



Conference fees



<ul><li>READ-COOP SCE members: free access</li><li>Undergraduates: 30 EUR</li><li>Regular attendees: 70 EUR</li></ul>



Coffee, lunch and conference dinner are included in the conference fee!



Program 



Thursday, 6th of February 



09:30-10:00 &nbsp;&nbsp;&nbsp;&nbsp;&nbsp; Registration



10:00-10:15 &nbsp;&nbsp;&nbsp;&nbsp;&nbsp; Opening words &#8211; Ulrike Tanzer (vice rector for research &#8211; University of Innsbruck) 



10:15-10:30 &nbsp;&nbsp;&nbsp;&nbsp;&nbsp; General introduction &#8211; Günter Mühlberger (University of Innsbruck)



10:30-11:00 &nbsp;&nbsp;&nbsp;&nbsp;&nbsp; After the grant: Transkribus as part of READ-COOP SCE. – Status quo and outlook&nbsp; &#8211; Andy Stauder (read-coop)



11:00-12:00 &nbsp;&nbsp;&nbsp;&nbsp;&nbsp; Transkribus in practice presentations



120 years of alpine mountaineering &#8211; the Alpenverein-project &#8211; Gerhard Rampl, Andrea Mussmann (University of Innsbruck)



Transkribus at the National Archives of the Netherlands &#8211; Liesbeth Keijser (National Archive Netherlands)



Ottoman Turkish &#8211; Suphan Kirmizialtin (University of New York)



ScanTent in practise &#8211; Erika Kuijpers (Vrije Universiteit Amsterdam)



Il Cartastorie &#8211; Banco di Napoli &#8211; Sergio Riolo, Andrea Zappulli (Banco di Napoli Foundation)



12:00-12:30 &nbsp;&nbsp;&nbsp;&nbsp;&nbsp; New features in Transkribus Sebastian Colutto (read-coop)



12:30-13:30 &nbsp;&nbsp;&nbsp;&nbsp;&nbsp; Lunch



13:30-15:00 &nbsp;&nbsp;&nbsp;&nbsp;&nbsp; Parallel workshops (1-3)



15:00-15:30 &nbsp;&nbsp;&nbsp;&nbsp;&nbsp; Coffee and cake



15:30-16:00 &nbsp;&nbsp;&nbsp;&nbsp;&nbsp; Share your model – public models in Transkribus



Achim Rabus (University of Freiburg)



Stefan Zathammer (University of Innsbruck)



Günter Hackl (University of Innsbruck)



Vincent Noppe (National Archives Netherlands)



16:00-16:30&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp; Search collections with Transkribus &#8211; Felix Dietrich, Berthold Ulreich (read-coop) 



16:30-17:30&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp; Parallel workshops (4-6)



from 20:00 &nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp; Dinner and come together at the &#8220;Biertstindl&#8221;-restaurant



Bierstindl: Klostergasse 6, 6020 Innsbruck (<a href="https://www.bierstindl.eu/">https://www.bierstindl.eu/</a>). We will organize meeting points for those who want to go by tram or to walk (30&#8242;):



<ul><li>Johanna at <a href="https://goo.gl/maps/QcCcSqgBdmAuAj5F7">Innsbruck Landesmuseum</a>, tram leaving at 19:33 (please ask in your hotel for tickets)</li><li>Günter walking from <a href="https://goo.gl/maps/mmFdBbEg5Er5YrM66">Annasäule </a>at 19:15</li></ul>



Friday, 7th of February 



09:30-10:30&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp; What users think about Transkribus &#8211; Melissa Terras (University of Edinburgh)



&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp; Crowd-sourcing with VeleHanden and Transkribus &#8211; Jirsi Reinders, Pauline van den Heuvel (Amsterdam City Archives) 



&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp; Parallel workshops (7-8)



10:30-11:00&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp; Transkribus in practice presentations



Late medieval writing &#8211; the Ambraser Heldenbuch &#8211; Aaron Tratter (University of Innsbruck)



Beyond 2022 &#8211; David Brown (Trinity College Dublin)



D.E.E.D.S-project &#8211; Hannah Lloyd (University of Toronto)



11:00-11:30&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp; Coffee and cake



11:30-12:30&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp; Visualizing data from parish registers &#8211; Herve Dejean (Naverlabs)



&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp; Cadastre maps with Transkribus &#8211; Günter Mühlberger (University of Innsbruck)



&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp; Parallel workshops (9-10)



12:30-13:00 &nbsp;&nbsp;&nbsp;&nbsp;&nbsp; Closing words



Workshops: 



Workshop 1: Transkribus for beginners &#8211; Tamara Terbul, Johanna Walcher (University of Innsbruck) 



Transkribus is an expert programme and it needs some training to fully understand the concept and all the features and possibilities offered by the programme. The WS will provide an introduction and enable users to work with their own documents.



Workshop 2: Training of large models &#8211; Dirk Alvermann (University of Greifswald), Achim Rabus (University of Freiburg) 



Many users ask for models which are able to read not only a specific hand, but are able to cope with many different writers. This WS will give some insights how such models can be created and especially how predictions can be made towards the expected quality of a model for unknown pages.



Workshop 3: Cadastre maps &#8211; Günter Mühlberger, Berthold Ulreich (University of Innsbruck) 



The first Austrian Cadastre was created during the 19th century. It consists of cadastre maps and accompanying documents. Within the WS we will give an introduction how the recognition of the cadastre maps is carried out and also how crowd users will be involved in the correction of tabular data.



Workshop 4: READ-COOP SCE Workshop &#8211; Daniel Wibmer (Raiffeisen Landesbank Tirol), Günter Mühlberger (University of Innsbruck) 



A European Cooperative Society is a powerful tool which provides excellent opportunities to collaborate on an intersectoral and international level. READ-COOP SCE is one of the first cooperatives in the cultural heritage and research domain. The WS will provide an opportunity to get a better understanding how cooperatives are working and how interested institutions and persons can take part to shape the future of Transkribus.



Workshop 5: Want to become a trainer for Transkribus? &#8211; Tamara Terbul, Johanna Walcher (University of Innsbruck), Annemieke Romein (University of Ghent), Stefano Bazzaco (University of Verona) 



The need for training courses and workshops for Transkribus is high and the team in Innsbruck receives many requests&nbsp; in this respect. However it is much more efficient if the Transkribus team needs not to travel but that training can be offered by local trainers. Therefore we are looking for people who would be interested and willing to become a Transkribus trainer. The WS will provide the opportunity to discuss this option and to hear about the experiences made already by people who organized Transkribus workshops independently from the Transkribus team.



Workshop 6: Working with tables &#8211; Herve Dejean (Naverlabs), Alexis Litvine (University of Cambridge) 



Tables are among the most important document types. Table recognition receives nowadays a lot of attention by research groups from all over the world.&nbsp; The WS will provide uptodate information on how to benefit from newest developments in the table recognition area.



Workshop 7: Structure recognition &#8211; Sebastian Colutto (read-coop), Günter Mühlberger (University of Innsbruck) 



With the P2PaLa tool it is possible to train structural features of an image. E.g. drawings can be separated from text, marginalia or &#8211; very useful &#8211; specific columns in tables can be detected automatically. The WS will give an introduction to the tool and show examples how it can be used in practise.



Workshop 8: Middle ages and early modern times &#8211; Tobias Hodel (University of Bern)



A significant amount of users in Transkribus are interested in documents from the middle ages and early modern times. Therefore already a large number of data are available and can be used to train robust general models. The WS is intended to gather all those who are specifically interested in this time period and who want to exchange their experiences with other users.



Workshop 9: Arabic, Hebrew and Osmanic in Transkribus (Suphan Kirmizialtin, University of New York) 



Several teams around the world have trained Arabic and Hebrew models in Transkribus. Even transliterations from Ottoman Turkish alphabet into modern Latin Turkish were performed with success. The workshop will provide background information on such projects and try to gather Transkribus users who are interested in this area.



Workshop 10: Text2Image &#8211; Günter Hackl (University of Innsbruck), Sebastian Colutto (read-coop) 



The t2i tool is useful for matching existing transcriptions with the a page image. It offers a manual mode as well as an semi-automated mode. In this way small as well as large collections of existing transcripts can be utilized for training models &#8211; without the need to carry out transcriptions from scratch.



Within the registration process you will be asked to choose the workshop sessions you prefer.



Accommodation information



We have contingents in three hotels, which you can check out and book under the following link:



For English-speakers: <a href="https://web4.deskline.net/innsbruck/en/accommodation/list?&amp;SelArrivalDate=20200205&amp;selNights=2&amp;lkPG=de63de3d-955f-4cfe-94a0-c561079014ff%20">https://web4.deskline.net/innsbruck/en/accommodation/list?&amp;SelArrivalDate=20200205&amp;selNights=2&amp;lkPG=de63de3d-955f-4cfe-94a0-c561079014ff </a>



For German-speakers: <a href="https://web4.deskline.net/innsbruck/de/accommodation/list?&amp;SelArrivalDate=20200205&amp;selNights=2&amp;lkPG=de63de3d-955f-4cfe-94a0-c561079014ff%0d">https://web4.deskline.net/innsbruck/de/accommodation/list?&amp;SelArrivalDate=20200205&amp;selNights=2&amp;lkPG=de63de3d-955f-4cfe-94a0-c561079014ff</a>



The contingents will be blocked until mid of December. If you would like to make use of this, please book as soon as you can as contingents are booked out quickly usually.



Alternatively, you can of course book your room directly at the hotel’s homepage or via a booking platform. Here are some more hotels we can recommend:



<ul><li>Austria Trend Hotel Congress: <a href="https://www.austria-trend.at/de/hotels/congress-innsbruck">https://www.austria-trend.at/de/hotels/congress-innsbruck</a></li><li>Best Western Hotel Goldener Adler: <a href="https://www.goldeneradler.com/">https://www.goldeneradler.com/</a></li><li>Hotel Innsbruck: <a href="https://www.hotelinnsbruck.com/">https://www.hotelinnsbruck.com/</a></li><li>Hotel Schwarzer Adler: <a href="https://www.schwarzeradler-innsbruck.com/home.html">https://www.schwarzeradler-innsbruck.com/home.html</a></li><li>The Penz: <a href="https://www.the-penz.com/">https://www.the-penz.com/</a></li><li>Stage 12: <a href="https://www.stage12.at/">https://www.stage12.at/</a></li></ul>



Open space for bilateral meetings on Friday afternoon



After the ending of the conference on Friday we will still have rooms available for spontaneous meetings between participants. If you have made a contact at the conference and would like to have a meeting before leaving, you can let us know in the course of the event, so we can provide a room to you.



Social events on Saturday 



For all those who will stay in Innsbruck for the weekend, we would like to organise two social events to choose. If you are interested, you can register via Conftool.



<ol><li>Skiing at Axamer Lizum </li></ol>



Axamer Lizum is a skiing resort in the vicinity of Innsbruck with possibility to go there by bus. For everybody who would like to join, we would meet directly in the skiing resort at 12 noon at the valley station of the “Schönbodenbahn” chairlift.



Via this link you can check out the skiing resort: <a href="https://www.axamer-lizum.at/en/service-our-guests.html">https://www.axamer-lizum.at/en/service-our-guests.html</a>



It is possible to borrow the skiing equipment directly at Axamer Lizum. All the info can be found here: <a href="https://www.axamer-lizum.at/en/skiing-school-amp-ski-rentals.html">https://www.axamer-lizum.at/en/skiing-school-amp-ski-rentals.html</a>



The price for a full day pass is € 41,50, but there are more options like half-day-tickets, tickets for special hours – all the prices for ski passes you can check out here: <a href="https://www.axamer-lizum.at/en/prices-axamer-lizum.html">https://www.axamer-lizum.at/en/prices-axamer-lizum.html</a>



The bus connection can be looked up here: <a href="https://www.vvt.at/page.cfm?vpath=index">https://www.vvt.at/page.cfm?vpath=index</a>



Stations to enter in the city centre of Innsbruck would be Innsbruck “Hauptbahnhof” or “Finanzamt”.



<ol><li>Sledging </li></ol>



Second option is sledging – destination will the Gleinserhöfe nearby Innsbruck. Together with Günter you will go there by bus and return by taxi &#8211; likely arrive in Innsbruck around 16:00. The climb will be about one and a half hour way and sledges can be borrowed for free. Before sledging down we will have Lunch together at the Gleinserhöfe.



We will meet at 10:45 at the <a href="https://goo.gl/maps/Ej3ecMjE8m69tuSH8">Main Railway Station &#8211; Bus terminal F</a>.



Bus 4140 to Schönberg/Stubaital. We will get out at Gemeindeamt Stubaital.



Please take good shoes (better boots), warm trousers, anorac, cap and&nbsp; gloves (important)!



Contact details



If you have any questions about the conference please contact&nbsp;<a href="mailto:email@transkribus.eu">email@transkribus.eu</a>



<figure class="wp-block-image"><img decoding="async" src="https://read.transkribus.eu/wp-content/uploads/2019/10/innsbruck-slider.jpg" alt=""/></figure>



<h6 class="wp-block-heading"><a href="https://read.transkribus.eu/wp-content/uploads/2019/10/innsbruck-slider.jpg"></a></h6>



<h6 class="wp-block-heading">Source: https://www.erklaervideo24.at/innsbruck.php</h6>

Transkribus User Conference 2020

We are happy to report about the Scanathon, which took place at the University of Greifswald as part of their open house day on Saturday, 7th of March 2020. The event was also connected to the German Archives Day (by the Verband deutscher Archivarinnen und Archivare) and besides the Scanathon itself, mini-workshops had been organised, giving participants the opportunity to get an insight into Transkribus.
The goal of the Scanathon was to digitise archival material with the help of volunteers, digitising with the ScanTent and subsequently make the documents available for scientific research via the Internet. About a thousand pages of the homeland-supplements of different Pomeranian Newspapers of the nineteen-twenties had been scanned in the course of the event. In the next step the team of the University Archive Greifswald uploaded the scans to Transkribus, where an automatic transcription had been generated.
The great result is that today 44 years of homeland supplements of Pomeranian Newspapers are available and searchable online. You can have a look at them via the digital library here: <a href="https://exchange.uibk.ac.at/owa/redir.aspx?C=OXyxDkdAeaRNPSGC4Crexih2CnKQf0SHXOdycontU--E5zeFT8rXCA..&amp;URL=https%3a%2f%2fdigitale-bibliothek-mv.de%2fviewer%2fbrowse%2f-%2f1%2f-%2fDC%3aarchive.100uaghgw.800zeitungen%2f">https://digitale-bibliothek-mv.de/viewer/browse/-/1/-/DC:archive.100uaghgw.800zeitungen/</a>
Want to read more about the event?<a href="https://www.uni-greifswald.de/universitaet/information/aktuelles/detail/n/scanathon-am-tag-der-offenen-tuer-im-universitaetsarchiv-greifswald-61123/"> https://www.uni-greifswald.de/universitaet/information/aktuelles/detail/n/scanathon-am-tag-der-offenen-tuer-im-universitaetsarchiv-greifswald-61123/</a>
More information about the University Archive Greifswald you can find here: <a href="https://www.uni-greifswald.de/en/university/facilities/archiv0/staff-members-at-the-university-archive/">https://www.uni-greifswald.de/en/university/facilities/archiv0/staff-members-at-the-university-archive/</a>
Many thanks to the participants and organisers for this fruitful event!
<a href="https://read.transkribus.eu/wp-content/uploads/2020/03/extract-Greifswald.jpg"><img decoding="async" class="wp-image-3614 aligncenter" src="https://read.transkribus.eu/wp-content/uploads/2020/03/extract-Greifswald.jpg" alt="" width="480" height="360" /></a><a href="https://read.transkribus.eu/wp-content/uploads/2020/03/scantent-logo.png"><img decoding="async" class="wp-image-3613 aligncenter" src="https://read.transkribus.eu/wp-content/uploads/2020/03/scantent-logo.png" alt="" width="135" height="105" /></a>
Image source: https://www.uni-greifswald.de/en/university/facilities/archiv0/

+ Scanathon at the University Archive Greifswald

On 6-7 February 2020 the third Transkribus User Conference took place at the Faculty of Catholic Theology in Innsbruck. 160 Transkribus users made their way to the snowy town of Innsbruck, which had been the conference’s venue for the first time.
Thanks to the great input of our invited speakers we managed to offer a versatile program and had the pleasure to get an insight into different Transkribus-related topics. The main items we had on our agenda were case studies, demonstrations of new tools and technological insights. All that with a “hands-on”-focus this year. To integrate the participants as much as possible we offered a total of ten workshops, covering topics like TextToImage, Structure Training, Tables and Cadastre maps.
In the “Transkribus in practice”-sessions experienced Transkribus users shared their knowledge about recognition projects, dealing with a variety of languages like Dutch, Italian and Ottoman Turkish. In the public model session engaged users were presenting their models, which recently became available for all Transkribus users. Models for Latin, Church Slavonic and Dutch had been amongst them.
We also heard about visualizing data from parish registers and about how the crowd-sourcing platform VeleHanden makes use of Transkribus, allowing volunteers to produce transcriptions that can then be automatically used as training data for recognising notarial documents.
As the READ-project, which was financed by the EU Horizon 2020 research and infrastructure program, has come to an end in summer 2019, we had a look into the future, where Transkribus will be further developed with the READ-COOP as legal entity. We are especially happy having been able to also welcome a good number of the COOP-members at the conference.
Currently our camera team is working on cutting the film material and as soon as possible we will make the presentation videos available to you. Transkribus users will be informed via email as soon as they will be publicised on Youtube.
The conference was accompanied by a couple of social events, which provided the opportunity to get in touch with other TUC-participants and exchange in a more casual setting, which was a great pleasure and lots of fun. We met at the pre-conference get-together at the Galway Pub on Wednesday evening to have some beer and wine in preparation for the conference.
After the conference day on Thursday we moved to the Bierstindl-restaurant for dinner. At the Tombola three people had been lucky and embraced their own ScanTent. Then it was time for our DJ to get started with an interesting song wish list (we had “Anton aus Tirol” of DJ Ötzi), which didn’t stop people from hitting the dance floor until the restaurant staff stopped them due to the curfew.
On Saturday we went sledging and skiing with the outdoor enthusiastic part of the TUC-participants, in order to enjoy the fresh snow. The weather was great, food at the mountain hut delicious and luckily we could send everybody home without serious injuries ?
We would like to thank all participants of the Transkribus User Conference again for making the event a great one. It was a place of informative presentations, inspiring talks and discussions and last but not least a lot of fun.
<a href="https://s3.eu-central-1.amazonaws.com/readcoop.cis.public-assets.prod/WP/uploads/2020/02/Pano-GruppeTUC2-scaled.jpg"><img decoding="async" class="alignnone wp-image-6263 size-large" src="https://s3.eu-central-1.amazonaws.com/readcoop.cis.public-assets.prod/WP/uploads/2020/02/Pano-GruppeTUC2-1024x500.jpg" alt="" width="1024" height="500" srcset="https://s3.eu-central-1.amazonaws.com/readcoop.cis.public-assets.prod/WP/uploads/2020/02/Pano-GruppeTUC2-1024x500.jpg 1024w, https://s3.eu-central-1.amazonaws.com/readcoop.cis.public-assets.prod/WP/uploads/2020/02/Pano-GruppeTUC2-scaled-600x293.jpg 600w, https://s3.eu-central-1.amazonaws.com/readcoop.cis.public-assets.prod/WP/uploads/2020/02/Pano-GruppeTUC2-300x146.jpg 300w, https://s3.eu-central-1.amazonaws.com/readcoop.cis.public-assets.prod/WP/uploads/2020/02/Pano-GruppeTUC2-768x375.jpg 768w, https://s3.eu-central-1.amazonaws.com/readcoop.cis.public-assets.prod/WP/uploads/2020/02/Pano-GruppeTUC2-1536x750.jpg 1536w, https://s3.eu-central-1.amazonaws.com/readcoop.cis.public-assets.prod/WP/uploads/2020/02/Pano-GruppeTUC2-2048x1000.jpg 2048w" sizes="(max-width: 1024px) 100vw, 1024px" /></a>
<a href="https://s3.eu-central-1.amazonaws.com/readcoop.cis.public-assets.prod/WP/uploads/2020/02/P2060079bea-scaled.jpg"><img decoding="async" class="alignnone wp-image-6262 size-large" src="https://s3.eu-central-1.amazonaws.com/readcoop.cis.public-assets.prod/WP/uploads/2020/02/P2060079bea-1024x768.jpg" alt="" width="1024" height="768" srcset="https://s3.eu-central-1.amazonaws.com/readcoop.cis.public-assets.prod/WP/uploads/2020/02/P2060079bea-1024x768.jpg 1024w, https://s3.eu-central-1.amazonaws.com/readcoop.cis.public-assets.prod/WP/uploads/2020/02/P2060079bea-scaled-600x450.jpg 600w, https://s3.eu-central-1.amazonaws.com/readcoop.cis.public-assets.prod/WP/uploads/2020/02/P2060079bea-300x225.jpg 300w, https://s3.eu-central-1.amazonaws.com/readcoop.cis.public-assets.prod/WP/uploads/2020/02/P2060079bea-768x576.jpg 768w, https://s3.eu-central-1.amazonaws.com/readcoop.cis.public-assets.prod/WP/uploads/2020/02/P2060079bea-1536x1152.jpg 1536w, https://s3.eu-central-1.amazonaws.com/readcoop.cis.public-assets.prod/WP/uploads/2020/02/P2060079bea-2048x1536.jpg 2048w" sizes="(max-width: 1024px) 100vw, 1024px" /></a>
<a href="https://s3.eu-central-1.amazonaws.com/readcoop.cis.public-assets.prod/WP/uploads/2020/02/sledding_05.jpg"><img decoding="async" class="alignnone wp-image-6261 size-large" src="https://s3.eu-central-1.amazonaws.com/readcoop.cis.public-assets.prod/WP/uploads/2020/02/sledding_05-1024x768.jpg" alt="" width="1024" height="768" srcset="https://s3.eu-central-1.amazonaws.com/readcoop.cis.public-assets.prod/WP/uploads/2020/02/sledding_05-1024x768.jpg 1024w, https://s3.eu-central-1.amazonaws.com/readcoop.cis.public-assets.prod/WP/uploads/2020/02/sledding_05-600x450.jpg 600w, https://s3.eu-central-1.amazonaws.com/readcoop.cis.public-assets.prod/WP/uploads/2020/02/sledding_05-300x225.jpg 300w, https://s3.eu-central-1.amazonaws.com/readcoop.cis.public-assets.prod/WP/uploads/2020/02/sledding_05-768x576.jpg 768w, https://s3.eu-central-1.amazonaws.com/readcoop.cis.public-assets.prod/WP/uploads/2020/02/sledding_05.jpg 1440w" sizes="(max-width: 1024px) 100vw, 1024px" /></a>

+ Transkribus User Conference 2020 – we keep transcribing as READ-COOP

From the 1930s until 1953 one veteran of the Irish Revolution (1912-23), Ernie O’Malley, interviewed over 450 of his former comrades about their experiences. Although deposited in UCD Archives since the 1970s, those notebooks have been largely inaccessible to scholars principally because of Ernie O’Malley’s impenetrable handwriting. As a result, one of the richest sources on the life and afterlife of the Irish Revolution has been underused and largely overshadowed by other accounts.
Because O’Malley’s uniquely gritty, graphic and atmospheric interviews will enhance the understanding of revolutionary and post-revolutionary Ireland, and because those notebooks also provide excellent sources for scholars engaged in wider comparative studies of political violence and civil war, a team of two editors, Dr Anne Dolan, Associate Professor in Modern Irish History, Trinity College Dublin, and Dr Eve Morrison, Canon Murray Fellow in Irish History at the University of Oxford, in collaboration with UCD Archives, decided to produce the first full scholarly edition of the notebooks.
That’s what they say about their work with Transkribus:
“Faced with a daunting, decade-long undertaking, we consulted our colleague Dr David Brown (Trinity College Dublin) about Transkribus. He has used Transkribus to marvellous effect on early modern manuscripts. If Transkribus could tackle O’Malley’s hand it would fundamentally change the prospects for our work, and it did.&nbsp; Our first model produced results beyond our wildest expectations.&nbsp; Transkribus transcribed over 120 notebooks in less than two weeks, saving us several years of work. We now have a full first draft of c.3,000,000 words to edit and prepare for publication.
The success we have had with Transkribus has allowed us to expand the scope of our project.&nbsp; As well making the transcriptions freely available online alongside digital copies of the notebooks through the UCD Digital Library the Irish Manuscripts Commission will also publish a full annotated edition over multiple volumes.&nbsp; Our use of Transkribus means that this invaluable source for the history of modern Ireland will no longer go overlooked.”
On this site you can find more information about projects of the UCD-Archives: <a href="https://exchange.uibk.ac.at/owa/redir.aspx?C=n-m9v0bU_BG-G_Ib6o5_BlnVYgXJ8rsUAac9dfm2GPXaSKbr1pjXCA..&amp;URL=http%3a%2f%2fwww.ucd.ie%2farchives">http://www.ucd.ie/archives</a>

+ Ernie O’Malley Interviews Project’s about Transkribus

We are happy to present one of our public models, which is the “Noscemus GM v1”-model released by Stefan Zathammer as part of the Innsbruck based project NOSCEMUS (Nova Scientia: Early Modern Scientific Literature and Latin). This model can read texts set in Antiqua-based typefaces from the 16th, 17th&nbsp;and 18th&nbsp;century, outperforming most standard OCR engines. Although it is tailored towards transcribing (Neo-)Latin texts, it provides convincing results also for other languages such as French, Italian or English. The Noscemus model can therefore not only provide help for Neo-Latinists, but for all kinds of research dealing with big text corpora from the Early Modern Period.



The model is based on training data coming from the Digital Sourcebook of the project and comprises about 1,000 pages. In order to keep the model as flexible as possible, standardizations in the transcription process were kept to a minimum. Only in the following cases normalizations were made: ligatures (e.g.&nbsp;ae,&nbsp;oe,&nbsp;ct,&nbsp;ff) and abbreviations (e.g.&nbsp;-que,&nbsp;-us,&nbsp;-tur,&nbsp;…mm…) were expanded, long s (ſ) was transcribed as normal&nbsp;s,&nbsp;small caps were transcribed as majuscules.



Even though the model provides already good results, the project is still dealing with a few issues: there are some remaining inconsistencies in the transcription of quotation marks and the error rate for the transcription of Greek words or passages is still high, to a smaller degree the same applies to (German) Fraktur.



We hope that the Noscemus-model will make transcription-life easier for many of you and for all those working on different kinds of documents, don’t forget to have a look at the other models we have been able to publicize recently thanks to our hard-working users. An overview about all our public models you can find in this document:&nbsp;<a href="https://transkribus.eu/wiki/images/d/d6/Public_Models_in_Transkribus.pdf">https://transkribus.eu/wiki/images/d/d6/Public_Models_in_Transkribus.pdf</a>

+ Presenting the Noscemus-public-model

We are excited to announce that new models had been made available to the Transkribus crowd recently. Getting closer to our goal of being a platform for model sharing and thanks to our hard-working users a total of 17 models is free for public use so far. These models cover different epochs and languages like English, French, German, Italian, Dutch and Devanagari. Via this link you can access the file with an overview of the models. Also you can learn which institution or person has created the model and find a description, which documents it is suitable for: <a href="https://readcoop.eu/transkribus/public-models/">https://readcoop.eu/transkribus/public-models/</a>
If you are keen on sharing a model yourself, please get in touch and we will check the details (<a href="mailto:info@readcoop.eu">info@readcoop.eu</a>). In case the data your model is trained on is sensitive and can not be published, we can take care that only the neural network is made available to other users, but the data is hidden. This way the model can “only” be used and no data behind is shown.
We are happy about seeing the Transkribus platform and community grow and are stoked about the great work our users are doing. Thanks again for your effort and keep transcribing…
<a href="https://read.transkribus.eu/wp-content/uploads/2019/12/Public-models.jpg"><img decoding="async" class="wp-image-3192 aligncenter" src="https://read.transkribus.eu/wp-content/uploads/2019/12/Public-models.jpg" alt="" width="657" height="188" /></a>

+ Public models: try out new ones and publicise your own!

The READ-COOP SCE with limited liability is now formally established! Under the identification number FN 520187g it is registered in the Austrian company register. READ-COOP SCE will run and further develop the Transkribus platform and related services such as the ScanTent. We are convinced that this governance model will give us plenty of opportunities to serve the needs of our members in a highly effective, open and democratic way. READ-COOP SCE is led by the board of directors consisting of Günter Mühlberger (chair), Andy Stauder (managing director) and Melissa Terras (director). The final say is with the General meeting of all members. Currently already more than 30 institutions and private persons have joined our coop. Everyone interested in the digitisation, transcription, text recognition, and searching of historical documents is wellcome to take part and to become a member! <a href="https://readcoop.eu/about/">More information</a>.

+ READ-COOP SCE formally established!

The Transkribus team collaborates with the Pattern Recognition team of the University Erlangen-Nürnberg (also member of READ-COOP SCE) and the collegues were so great to make an interesting experiment: to train their classifier for discriminating printed and handwritten text lines automatically. There are mainly two use-cases: (1) to improve recognition results if specific HTR models are applied to specific script types. However, we made the experience that the HTR engines usually can deal rather well with a large number of scripts internally so the actual benefit may not be as high as one expects. (2) to find handwritten lines in printed books. E.g. if famous persons made notes in their private books the tool which is described below will find them with amazing accuracy!
The following text was provided by Matthias Seuret and Vincent Christlein from the <a href="https://www5.cs.fau.de/nc/en/our-team/">Pattern Recognition team</a> and slightly adapted for this post:
The difficulty in the classification of text lines as being printed or handwritten does not lie much in the usage of the convolutional neural networks (CNN) or the design of their architecture, but in the acquisition and preparation of the data. Indeed, modern artificial neural networks (ANN) are now able to deal with highly complex data (such as ImageNet, which includes 90 different dog breeds to discriminate), and for a large variety of tasks, presenting enough examples to the ANN is sufficient to make it reach a fair accuracy.
It is necessary to note that ANNs (and other artificial intelligence systems) are extremely biased by the data used to train them. Because of this, the training data should be chosen carefully to ensure that the easiest way to classify the images correctly is to solve the task. For example, if all (or most) handwritten text lines are on a yellowish paper, while printed material is on white paper, then the ANN will simply learn to separate yellow from white, and will answer that any text line printed on yellowish paper is handwritten. Of course, an ANN can learn various other undesired data properties, such as the image resolution and quality, the texture of the paper, or the colour or contrast of the ink. Thus, it is of an utmost importance to use training data as similar as possible to what the ANN will have to deal with.
The system we developed for this task is based on the (printed) font groups classifier developed for the OCR-D project (<a class="moz-txt-link-freetext" href="http://www.ocr-d.de/">http://www.ocr-d.de/</a>). It consists of a DenseNet-121 wrapped in some utility classes, and has been adapted for the binary classification of handwritten and printed text. The DenseNet-121 is a convolutional neural network with 121 layers, most of them being in 4 blocks densely connected. It has however a relatively small amount of parameters for a network of its size, and thus requires less data to be trained than architectures with more parameters.



<figure class="wp-block-image"><img decoding="async" src="https://read.transkribus.eu/wp-content/uploads/2019/09/nsmail-6.png" alt="" class="wp-image-2885"/><figcaption>Machine learning schema for printed vs. handwritten text lines</figcaption></figure>



Text lines are pre-processed in two ways. First, all of them are resized to a height of 150 pixels, and their aspect ratio is preserved. This is helpful for the ANN, as it will not have to learn dealing with a large diversity of text size. Second, data augmentation methods applied to the training images. This means that some small modifications, such as shearing or hue modification, are applied to the training images every time they are shown to the neural network during the training. The goal is to make the network learn to ignore these variations and perform on unseen data.



We trained our network on text lines coming from two different sources. Approximately 40&#8217;000 printed text lines were extracted automatically from the dataset presented in &#8220;Dataset of Pages from Early Printed Books with Multiple Font Groups&#8221; (<a href="https://doi.org/10.5281/zenodo.3366685">https://doi.org/10.5281/zenodo.3366685</a>), and 9&#8217;577 handwritten samples provided by READ. Also, 1&#8217;562 text lines from each class were used for test purpose &#8211; none of them came from a page used for the training data. While our network reached a classification accuracy of 97.5% on the test data, one has to keep in mind that this holds true only for this specific data. The source code of our method and the trained CNN, as well as code allowing anybody to easily retrain the CNN on their own data, is available at the following address: <a href="https://github.com/seuretm/printed-vs-handwritten">https://github.com/seuretm/printed-vs-handwritten</a>



Note: If you are interested to create training data for this purpose in Transkribus you can use the &#8220;Structural tagging&#8221; feature and mark lines as &#8220;handwritten&#8221; or &#8220;printed&#8221; in your documents. The actual classifier needs to run outside of Transkribus, however, if there is some strong support from the user community we are happy to include the tool also in the Transkribus platform.

+ Printed vs. handwritten text lines &#8211; automatically separated

Dr. Dirk Alvermann of the University Archive Greifswald is one of the pioneers of Transkribus. He already started working with the first version of Transkribus in 2015. Now, he received a grant from the German Science Funds (DFG) for digitising and recognizing about 240,000 pages of court records from the 16th to 18th century. Moreover the University of Greifswald is one of the founding members of READ-COOP SCE.
Dirk Alverman and his team recently put online a blog, which offers explanations on different Transkribus-topics. By having a look at the blog, you will not only get a good understanding of things like reading order, ground truth or layout analysis in Transkribus, but also read about how to use it on your documents. As this project team has already been working with Transkribus for some time, members are able to give an insight into the practical aspects of digitisation projects processed with the help of Transkribus.
Became curious? Have a look at the blog here: <a href="https://rechtsprechung-im-ostseeraum.archiv.uni-greifswald.de/">https://rechtsprechung-im-ostseeraum.archiv.uni-greifswald.de/</a>
&nbsp;
<a href="https://read.transkribus.eu/wp-content/uploads/2019/09/Archiv-Greifswald.jpg"><img decoding="async" class="wp-image-2876 aligncenter" src="https://read.transkribus.eu/wp-content/uploads/2019/09/Archiv-Greifswald.jpg" alt="" width="488" height="244" /></a>Source: <a href="https://www.uni-greifswald.de/universitaet/organisation/verwaltung/stabsstelle-universitaetsarchivkustodie/">https://www.uni-greifswald.de/universitaet/organisation/verwaltung/stabsstelle-universitaetsarchivkustodie/</a>
&nbsp;

+ Digitisation blog of the University Archive Greifswald

A general paper about Transkribus was published in the Journal of Documentation.  <a href="https://www.emerald.com/insight/content/doi/10.1108/JD-07-2018-0114/full/pdf?title=transforming-scholarship-in-the-archives-through-handwritten-text-recognition-transkribus-as-a-case-study">Transforming scholarship in the archives through handwritten text recognition</a> gives an overview of the current use of HTR on archival manuscript material with the help of Transkribus. Several use cases are discussed and therefore the article demonstrates how HTR can affect scholarship and revolutionize the use of digitised heritage content.
The article shows that HTR is an efficient technology, suitable for the work on mass digitisation material. Cooperative models are a further way of sustainable use of the Transkribus platform. The fact, that Transkribus is the only publicly available platform for HTR involves the potential to be path-breaking for institutions as well as individuals by providing information, which wasn’t accessible before. Important note: The article was written in 2018 and therefore figures for recognition rates have improved significantly in the meantime!
Have a look at the article, it will give you an overview about the current use of HTR within a wide archival studies community. Via this link you can access it: <a href="https://www.emerald.com/insight/content/doi/10.1108/JD-07-2018-0114/full/html">https://www.emerald.com/insight/content/doi/10.1108/JD-07-2018-0114/full/html</a>
<a href="https://read.transkribus.eu/wp-content/uploads/2019/09/old-books-436498_1920.jpg"><img decoding="async" class="wp-image-2870 aligncenter" src="https://read.transkribus.eu/wp-content/uploads/2019/09/old-books-436498_1920.jpg" alt="" width="563" height="375" /></a>Source: <a href="https://pixabay.com/de/photos/alte-b%C3%BCcher-buch-alte-bibliothek-436498/">https://pixabay.com/de/photos/alte-b%C3%BCcher-buch-alte-bibliothek-436498/</a>

+ Paper on Transkribus and handwritten text recognition (HTR) in archives now open access

On 27th of August Barbara Denicolò of the Transkribus Team Innsbruck held a workshop at the summer school &#8220;Digital Methods in Humanities and Social Sciences&#8221; at the University of Tartu (Estonia). An overview about different analysis methods and methodological principles for working with (digital) data in humanities and social sciences was given within the framework of this summer school. Transkribus was invited to represent the historical-paleographical part, i.e. the work with handwritten and printed text.
The workshop met with great interest and was booked out quickly. Half of the participants were PhD students, additionally professors, postgraduates and independent (senior) researchers took part. The three-part workshop, which lasted throughout the day, began with a detailed and wide-ranging presentation of the program, its history and its various features, as well as a brief theoretical presentation of an exemplary workflow. After the lunch break, people were working on their manuscripts directly. Exercise examples in different languages and screens were available for them in order to discuss individual project requirements.
More information about the &#8220;Digital Methods in Humanities and Social Sciences&#8221; summer school in Tartu can be found here:
<a href="https://digitalmethods.ut.ee/node/19198">https://digitalmethods.ut.ee/node/19198</a>
&nbsp;
<a href="https://read.transkribus.eu/wp-content/uploads/2019/09/Tartu-University.jpg"><img decoding="async" class="wp-image-2862 aligncenter" src="https://read.transkribus.eu/wp-content/uploads/2019/09/Tartu-University.jpg" alt="" width="646" height="323"></a>Source: http://www.mladiinfo.eu/2017/12/15/degree-programmes-with-tuition-waiver-scholarship-at-the-university-of-tartu/

+ Successful Transkribus Workshop at DH summer school in Tartu

The SS Great Britain trust accepted the challenge of deciphering Isambard Kingdom Brunel’s handwriting. Without technical help this had been a challenge. The team in charge discovered, that his handwriting was “almost impossible to read”. That’s where Transkribus came into play: even though the project-team only started to use Transkribs and the amount of training data wasn’t very high, useful results had already been achieved. Still there is a lot of work to do, as there are several thousand pages of Brunel’s handwriting.
&#8220;It&#8217;s hugely sped up the process and we&#8217;re learning new bits about his life; there&#8217;s so much potential to unlock”, says Nick Booth from the SS Great Britain trust. The SS Great Britain in Bristol is keeping the collection of material about the life and works of Brunel and tells the story of one of Britain’s greatest engineers and Brunel’s SS Great Britain, one of the most important historic ships in the maritime history.
The BBC news became aware of the project and published an article about it, which you can find here: <a href="https://www.bbc.com/news/uk-england-bristol-49347472">https://www.bbc.com/news/uk-england-bristol-49347472</a>
More information about Brunel and the SS Great Britain can be found on this homepage: <a href="https://www.ssgreatbritain.org">https://www.ssgreatbritain.org</a>
<a href="https://read.transkribus.eu/wp-content/uploads/2019/08/Brunel.png"><img decoding="async" class="size-full wp-image-2847 aligncenter" src="https://read.transkribus.eu/wp-content/uploads/2019/08/Brunel.png" alt="" width="764" height="440" /></a>
Source: https://www.bbc.com/news/uk-england-bristol-49347472

+ Decoding famous British engineer’s handwriting

The digitisation team around Liesbeth Keyser from the <a href="https://www.nationaalarchief.nl/">National Archives</a> in the Netherlands is working hard on creating training data for their collections in order to prepare HTR processing on a large scale. As a first result a model based on 475.769 words is now made available for Transkribus users. The model shows a Character Error Rate of 7.48% on the training set and 6.15% on the validation set. It is based on the careful transcription of dozens of different handwritings and comprises scans from the Incoming Documents from the Dutch East India Company (Overgekomen Brieven en Papieren van de VOC) of the National Archives of the Netherlands and of 19th century Notarial deeds from the Noord-Hollands archief.  The model is named: NAN/NHA_GT_M3+ Enjoy!
<div> </div>
&nbsp;

+ National Archives releases first version of a Dutch handwriting model

<a href="https://www.slavistik.uni-freiburg.de/personal/univ-prof-dr-achim-rabus">Prof. Achim Rabus</a> from the University of Freiburg has released two specialized models which are able to read Russian Curch Slavonic. The first model is called VMC_Test_4+: Training data consist of parts of the Russian Church Slavonic Great Reading Menology (16th century). The model is tailored towards transcribing Cyrillic semi-uncial script from the 16th century. Character Error Rates for the training data are 3.72% and for the validation set 3.92% and for the validation set 3.82%.
The second model is called: Combined_Full_VKS_2: Training data consist of parts of the Russian Church Slavonic Great Reading Menology (16th century), Old Church Slavonic Codex Suprasliensis (11th century), and the 11th century manuscript of the Catecheses of Cyril of Jerusalem. This is a generic model suitable for transcribing a variety of Old Cyrillic script styles including uncial and semi-uncial. Character Error Rates for the training data are 4.42% and for the validation set 3.92%.
Achim has written a <a href="https://www.academia.edu/38835297/Recognizing_handwritten_text_in_Slavic_manuscripts_A_neural-network_approach_using_Transkribus_1_Achim_Rabus">detailed report</a>  about his usage of Transkribus which is an excellent example how such a general model can be created. Thanks a lot!
&nbsp;

+ Special models on Slavic handwriting released

Thanks to the <a href="https://labs.onb.ac.at/">Library Labs</a> of the <a href="https://onb.ac.at/">Austrian National Library</a> and the <a href="http://newseye.eu/">NewsEye</a> project we are happy to announce the release of a free model which is capable to read German Fraktur documents especially from the 19th and 20th century in a convincing quality outperforming most standard OCR engines. The model is based on training data coming from the <a href="http://anno.onb.ac.at/">ANNO collection</a> of the Austrian National Library and comprises 442.141 words. It shows a CER of 1,55% on the training set and 1,65% on the test set without any dictionary support. Note: the model is trained on German language documents. It will provide less convincing results for other languages, such as Swedish or Finnish Fraktur. However models for these languages are also in preparation and may be released in the coming months. The Fraktur model is available for every registered user in Transkribus and called: ONB _Newseye_GT_M1+. Have fun!

+ General model for &#8220;Fraktur&#8221; released

We are excited about our next Transkribus User Conference, which this time will take place in Innsbruck. Preparations are in full swing and the date will be announced within the next few weeks.
Our last Transkribus User Conference in November 2018 was a great success, but there is always room for improvement. If you have some input to share on topics for the next conference or other suggestions for improvement, we are happy to receive them via <a href="mailto:email@transkribus.eu">email@transkribus.eu</a>.
The program of our last user conference can be found here: <a href="https://read.transkribus.eu/transkribus-user-conference-2018/">https://read.transkribus.eu/transkribus-user-conference-2018/</a>
… videos of the presentations in our Youtube-channel: <a href="https://www.youtube.com/playlist?list=PL7UbQtd4qlhKCEgLnZbJKQu9qpA5iF-sC">https://www.youtube.com/playlist?list=PL7UbQtd4qlhKCEgLnZbJKQu9qpA5iF-sC</a>
Hope to meet you there!

+ In preparation for the next Transkribus User Conference

Members of the Zurich University compared two versions of the ABBYY FineReader (FineReader XIX and FineReader Server 11) OCR (Optical Character Recognition) and the Transkribus HTR (Handwritten Text Recognition) in order to find out which one is the most effective one when it comes to recognition results on black letters in historical newspapers. For the test they used PDFs with medium resolution images of the German-language Neue Zürcher Zeitung.
The recognition of black letters in historical newspapers can be particularly challenging because the distinctiveness of characters is often low, the paper quality can be bad and, in many cases, small font sizes are used. Systems like ABBYY FineReader and Transkribus are working on tackling such problems. We are happy that the experiment of the University of Zurich shows that Transkribus provides significantly better results than the commercial system ABBYY FineReader.
The article explains the effectiveness of the HTR, as only a modest amount of manual work is needed for the creation of ground truth, which makes it possible to apply the HTR on documents. Especially with printed texts in newspapers, error rates in Transkribus are usually low. Moreover, the test shows that the model, which had been trained for the Neue Zürcher Zeitung, also provided good results for other newspapers of the same epoch, like the Bundesblatt and the Neue Zuger Zeitung. Good news is, that the model of the Neue Zürcher Zeitung will become public during 2019.
If you would like to have a closer look on the experiment, you can find the whole article here: <a href="https://dev.clariah.nl/files/dh2019/boa/0694.html">https://dev.clariah.nl/files/dh2019/boa/0694.html</a>
<a href="https://read.transkribus.eu/wp-content/uploads/2019/07/Grafik-Utrecht-University.png"><img decoding="async" class="alignnone size-full wp-image-2807" src="https://read.transkribus.eu/wp-content/uploads/2019/07/Grafik-Utrecht-University.png" alt="" width="1182" height="766" /></a>
Source: https://dev.clariah.nl/files/dh2019/boa/0694.html

+ Transkribus HTR competing in OCR-test of the Zurich University

On 1st of July 2019 the READ project will turn into a European Cooperative Society (SCE). READ-COOP will serve as the basis for sustaining and further developing the Transkribus platform and related services and tools.
READ-COOP will be based on the EU directive of a <a href="https://ec.europa.eu/growth/sectors/social-economy/cooperatives/european-cooperative-society_en" target="_blank" rel="noopener noreferrer">European Cooperative (SCE)</a>. Though the SCE will be set up according to EU law it will be open to members outside of the European Community as well. If you are interested in working with Transkribus on a long run &#8211; join READ-COOP and benefit from the work done by your collaborators.
One of the main reasons that we decided to go for a coop is that we want to support a &#8220;culture of collaboration&#8221; between archives/libraries, humanities scholars, computer scientists and the public (volunteers). We believe that intersectoral collaboration and full control over data are key for a successful integration of machine learning technologies into society and daily life. And an SCE delivers the best infrastructure to realize this goal.
An SCE is a legal entity which is open to new members (institutions, natural persons). Members shall benefit from an SCE directly, there is no shareholder value. Moroever SCEs are organised in a democratic way: The final say has the General Meeting.
More information can be found here: <a href="https://read.transkribus.eu/coop/">https://read.transkribus.eu/coop/</a>
&nbsp;

+ Foundation of READ-COOP

In May, Barbara Denicolò from the Transkribus-Team Innsbruck and Elena Mühlbauer from the Diözesanarchiv in Passau in the name of READ travelled to the Midwest of the USA to present Transkribus to the American audience.
Though small, Kalamazoo in the state of Michigan, is well-known for one of the major congresses of the various mediaeval disciplines, which takes place every year at the Western Michigan University WMU. At the 54th International Congress on Medieval Studies, Elena and Barbara presented Transkribus as a practical tool for philologists and historians to transcribe and annotate old manuscripts and prints manually or automatically.
After a brief general introduction, the participants were able to segment and transcribe various documents themselves in a test collection created especially for them in a total of 90 minutes, and to apply particularly good models themselves. Although the workshop was unfortunately scheduled for the last time slot on Sunday and there were already noticeably less congress participants walking across the large university campus, a good dozen interested people found their way to the workshop.
The great interest was noticeable, while participants talked about their own projects and discussed possible applications and use cases. Hanna Lloyd from the University of Toronto for example reported on her user experiences and research results in her lecture &#8220;Digitizing Paleography: Transcribing Latin Charters with Transkribus&#8221;.
More information about the conference can be found here:
<a href="https://wmich.edu/sites/default/files/attachments/u434/2019/medieval-congress-program-2019.pdf">https://wmich.edu/sites/default/files/attachments/u434/2019/medieval-congress-program-2019.pdf</a>
<a href="https://wmich.edu/medievalcongress">https://wmich.edu/medievalcongress</a>
<a href="https://read.transkribus.eu/wp-content/uploads/2019/06/Barbara_Elena_US.jpg"><img decoding="async" class="size-full wp-image-2779 aligncenter" src="https://read.transkribus.eu/wp-content/uploads/2019/06/Barbara_Elena_US.jpg" alt="" width="496" height="375"></a>

+ Transkribus goes America

In the past 3 years research groups and archives from all over Europe were working on Handwritten Text Recognition for historical documents. Results can now be seen at the public Transkribus seminar at the National Archives of Finland in Helsinki on Wednesday 26.6.2019!
The Transkribus platform enables non-technical users to train neural networks in order to recognize and search historical documents. The seminar will provide an update on latest technical developments and showcase how Transkribus can be used in various scenarios. Moreover, a first version of a web-interface for searching Finnish Court records from the 19th century will be launched. With this search interface users can search historical documents in a “Google like” way.
The READ project is currently in the transformation to become one of the first European Cooperative Societies in the research, education and cultural heritage domain. Institutions and private persons are warmly invited to join this initiative.
If you would like to take part, please register yourself under the following link (participation is free of charge and registration is possible until 18.6.2019): <a href="https://exchange.uibk.ac.at/owa/redir.aspx?C=Z6Zppnc4fILbbYvp9j2uoGR_ooiGECvaVmjGQjcV1cOs63gRS-rWCA..&amp;URL=https%3a%2f%2fwww.eventbrite.com%2fe%2ftranskribus-seminar-at-the-national-archives-of-finland-tickets-61567839064">https://www.eventbrite.com/e/transkribus-seminar-at-the-national-archives-of-finland-tickets-61567839064</a>
The program includes an inspiring set of presentations from our international partners, as well as lunch and a panel discussion:
10.00. Welcoming words
10.15. READ-COOP: Günter Mühlberger (UIBK)
      Transkribus and the technology behind it
10.45. Transkribus platform: Sebastian Colutto (UIBK)
11.15. HTR in READ and Transkribus: Roger Labahn, Gundram Leifert (URO and CITlab)
11.30. Segmentation tools: Sofia Ares Oliveira (EPFL)
11.45. Table recognition: Hervé Déjean (NAVER)
12.00. ScanTent and DocScan: Matthias Wödlinger (CVL)
12.15- 13.15 Lunch
      Transkribus in practice
13.15 <a href="https://exchange.uibk.ac.at/owa/redir.aspx?C=h3FdWvsXYxxGhYvurguIOkJgMuZeTDwObeJ4IoZqujes63gRS-rWCA..&amp;URL=http%3a%2f%2fedelfelt.sls.fi%2f">Edelfelt project</a>: Maria Vainio-Kurtakko (SLS)
13.45 <a href="https://exchange.uibk.ac.at/owa/redir.aspx?C=-2YhxptiA5xQjSgQqp-oKe_Y-aXKkmCClkXuRCIMkOus63gRS-rWCA..&amp;URL=https%3a%2f%2fvelehanden.nl%2f">VeleHanden</a>: Marc Ponte, Jirsi Reinders
14.15 <a href="https://exchange.uibk.ac.at/owa/redir.aspx?C=WHYp8y7SgIx_yBlYWZEel5RvvTgABevIM0lZH3PTN4Gs63gRS-rWCA..&amp;URL=https%3a%2f%2fmakingamodernarchive.blogspot.com%2f">Court Records Collection</a>: (NAF and UPVLC)
15.00. Panel discussion
<a href="https://read.transkribus.eu/wp-content/uploads/2019/06/helsinki-1269310_1280.jpg"><img decoding="async" class="alignnone size-full wp-image-2771" src="https://read.transkribus.eu/wp-content/uploads/2019/06/helsinki-1269310_1280.jpg" alt="" width="1280" height="720" /></a>
Source: https://pixabay.com/photos/helsinki-city-night-finland-1269310/

+ Handwritten Text Recognition at the National Archives of Finland

Recently our new HTR+ was tested on different styles of Church Slavonic handwritings by Achim Rabus, who is holding the Chair of Slavic Linguistics at the University of Freiburg in Germany. With Transkribus’ technology the error rates went down to 3 to 5 percent. Superscript letters, abbreviations and word separation are the challenges the HTR+ had to deal with.
A paper on the topic of recognizing handwritten text in Slavic manuscripts with Transkribus is about to be publicised by Achim Rabus. Within this project he discovered the potential of Transkribus when it comes to the digitizing of Church Slavonic manuscripts: the possibility to search in big documents without even having a special model for the individual handwriting and the opportunity to avoid a full manual transcription and instead just correcting the mistakes of the automated transcription makes “digitisation-life” a lot easier.
Part of the models Achim Rabus has trained already contain different hands and provide useful automatic transcripts. Nevertheless the READ-Team is working on further improving Transkribus in the way, that also for documents with mixed handwritings automatic transcripts with low character error rate can be produced.
Cooperation is the key for getting out the biggest benefit for everybody. That is also what Achim Rabus is convinced of and therefore he is happy to share his model with interested people. You can get in touch with him via email: <a href="mailto:achim.rabus@slavistik.uni-freiburg.de">achim.rabus@slavistik.uni-freiburg.de</a>
You can have a look at the draft of the paper Recognizing handwritten text in Slavic manuscripts: A neural-network approach using Transkribus under the following link: <a href="https://www.academia.edu/38835297/Recognizing_handwritten_text_in_Slavic_manuscripts_A_neural-network_approach_using_Transkribus_1_Achim_Rabus">https://www.academia.edu/38835297/Recognizing_handwritten_text_in_Slavic_manuscripts_A_neural-network_approach_using_Transkribus_1_Achim_Rabus</a>
<a href="https://read.transkribus.eu/wp-content/uploads/2019/05/Slavic-manuscript_Rabus.jpg"><img decoding="async" class="wp-image-2766 aligncenter" src="https://read.transkribus.eu/wp-content/uploads/2019/05/Slavic-manuscript_Rabus.jpg" alt="" width="579" height="200" /></a>Source: Rabus, Achim: Recognizing handwritten text in Slavic manuscripts: A neural-network approach using Transkribus 
&nbsp;

+ HTR+ reads old Slavonic documents with 3-5 % Character Error Rate

Transkribus now helps to produce a digital encyclopaedia, containing articles regarding sociability during the Age of British enlightenment. This should be achieved within the H2020 DIGITENS-project coordinated by the University of Western Brittany (UBO) in Brest, France, which gives young scholars the chance to get familiar with new digital humanities research tools and with the work in archives. At the same time the project opens them up to the opportunity to spend time abroad and therefore supports mobility.
For the DIGITENS-project, it is important to have a standardized workflow in order to work efficiently. This is where Transkribus comes into play. Our software makes it possible to cover the whole workflow from scanning with the ScanTent and the DocScan app up to international cooperation in using Handwritten Text Recognition models. This way Transkribus can give the project the required infrastructure.
To give the scholars an insight into the work with Transkribus, we have organised a workshop in Brest at the UBO on 22nd of May.
For more information about the project, visit the DIGITENS website: <a href="https://www.univ-brest.fr/digitens/">https://www.univ-brest.fr/digitens/</a>
The project is coordinated by the <a href="https://www.univ-brest.fr/gis-sociabilites">GIS Sociability</a> and the research lab <a href="https://www.univ-brest.fr/hcti">HCTI</a>. The DIGITENS encyclopedia will soon be available online at the following address: <a href="https://www.digitens.org/en">https://www.digitens.org/en </a>
<a href="https://read.transkribus.eu/wp-content/uploads/2019/04/Digitens.jpg"><img decoding="async" class="alignnone size-full wp-image-2744" src="https://read.transkribus.eu/wp-content/uploads/2019/04/Digitens.jpg" alt="" width="906" height="325" /></a>
http://www.digitens.fr/1/accueil

+ Transkribus-support for DIGITENS

On 4 April 2019 Transkribus users from the Netherlands and Belgium gathered under cloudy skies in The Hague to discuss the possibility of forming a network to improve the automated recognition of Dutch language documents.
The event was kindly hosted by Liesbeth Keijser and her team at <a href="https://www.nationaalarchief.nl/en" target="_blank" rel="noopener noreferrer">Nationaal Archief</a>.
The event attracted 45 people from more than 15 institutions including Nationaal Archief, Huygens ING, Koninklijke Bibliotheek, Stadsarchief Amsterdam, Europeana, International Institute of Social History, Ghent University, Het Utrechts Archief, Stadsarchief Antwerpen, Noord-Hollands Archief and Picturae.
There are many active Transkribus users in the Dutch-speaking regions and The Netherlands in particular is advanced in the realms of digitisation, technological innovation and digital humanities. This event was designed to allow users to share information about their work with Transkribus and forge collaboration on a generic &#8216;Dutch model&#8217; capable of recognising a sizeable variety of Dutch language documents.
We were welcomed with an introduction from Marens Engelhard, Director of Nationaal Archief.
The event moved on with a <a href="https://read.transkribus.eu/wp-content/uploads/2019/04/muehlberger_theHague.pdf">presentation about the vision for  a Dutch network from Günter Mühlberger, coordinator of the READ project.</a>
Günter explained that the launch of READ-COOP in the summer of 2019 will provide sustainability for Transkribus after the end of the READ project. He also offered a sneak peak of some forthcoming features in the platform including:
<ul>
<li>Advanced error rate tool to assess the accuracy of individual pages of automatically generated text</li>
<li>Trainable Layout Analysis capable of recognising page features like dates and marginalia</li>
<li>Improved interface for HTR model training, which makes it easier to mix different models together</li>
</ul>
We then heard from four sets of Transkribus users about the HTR models they had created and their experiences with the platform:
<ul>
<li style="font-weight: 400;"><a href="https://read.transkribus.eu/wp-content/uploads/2019/04/Keijser_Liakos-Nationaal-Archief.pdf">Liesbeth Keijser and Filotas Liakos (Nationaal Archief)</a></li>
<li style="font-weight: 400;"><a href="https://read.transkribus.eu/wp-content/uploads/2019/04/Ponte-Stadsarchief-Amsterdam.pdf">Mark Ponte (Stadsarchief Amsterdam)</a></li>
<li style="font-weight: 400;"><a href="https://read.transkribus.eu/wp-content/uploads/2019/04/Van-Rossum-IISG.pdf">Matthias van Rossum (International Institute of Social History)</a></li>
<li style="font-weight: 400;"><a href="https://read.transkribus.eu/wp-content/uploads/2019/04/Romein-Ghent-University.pdf">Christel Annemieke Romein (Ghent University)</a></li>
</ul>
<figure id="attachment_2726" aria-describedby="caption-attachment-2726" style="width: 3264px" class="wp-caption aligncenter"><a href="https://read.transkribus.eu/wp-content/uploads/2019/04/20190404_115135.jpg"><img decoding="async" class="wp-image-2726 size-full" src="https://read.transkribus.eu/wp-content/uploads/2019/04/20190404_115135.jpg" alt="" width="3264" height="1836" /></a><figcaption id="caption-attachment-2726" class="wp-caption-text">Liesbeth Keijser (Nationaal Archief) presenting</figcaption></figure>
The afternoon was dedicated to discussion about necessary milestones and next steps. Participants discussed the challenge of recognising Dutch material, possible avenues for exchanging data and the desirability of an individual coordinator to play a leading role in the network.
The idea of a Dutch language Transkribus network was enthusiastically received and the event concluded with suggestions of funding avenues to investigate and a proposal for twice-yearly meetings be held at different venues. There is huge potential here to work collaboratively to significantly improve the recognition of Dutch language material and we look forward to seeing what develops!

+ First meeting of Dutch Transkribus network

We are proud to be part of a successful project carried out by the <a href="https://alpineclub.org.nz">New Zealand Alpine Club</a> and the University of Innsbruck (Linguistic Institute). The complete workflow was done within Transkribus: apart from uploading files and running the text recognition volunteers used the web-based transcription interface from Transkribus to carefully correct all 17,500 pages of the New Zealand Alpine Journal.
In order to make the journal also searchable, Transkribus team members developed a simple but effective <a href="https://www.nzaj-archive.nz/">web-application</a>, which enables users to browse all issues and to search the full-text of the complete journal. The application runs also very well on a smartphone. All data is hosted by Transkribus.
The project received its funding via the crowd-funding-platform <a href="https://givealittle.co.nz/cause/accessible-digital-nz-alpine-journal-1892-2012">Give a little</a>. People donated about 6 000 NZD, which is a great support.
<a href="https://www.nzaj-archive.nz/">Check out the web-application for searching the journal editions here.</a>
&nbsp;
<a href="https://read.transkribus.eu/wp-content/uploads/2019/04/Alpine-Journal-Index.jpg"><img decoding="async" class="wp-image-2722 aligncenter" src="https://read.transkribus.eu/wp-content/uploads/2019/04/Alpine-Journal-Index.jpg" alt="" width="551" height="440" /></a>
Foto credit: https://www.nzaj-archive.nz/

+ Searching more than 100 years of mountaineering history with Transkribus

When we work together, there&#8217;s so much we can achieve! <a href="https://www.amsterdam.nl/stadsarchief/" target="_blank" rel="noopener noreferrer">Amsterdam City Archives</a> and <a href="https://velehanden.nl/" target="_blank" rel="noopener noreferrer">VeleHanden</a> have just launched a fantastic crowdsourcing initiative which combines the power of our Handwritten Text Recognition (HTR) technology with the talents of volunteer transcribers.
<ul>
<li>Try out <a href="https://velehanden.nl/projecten/bekijk/details/project/amsterdam_correct_notarieel_transkribus" target="_blank" rel="noopener noreferrer">Crowd leert computer lezen</a> (or Crowd teaches the computer to read)</li>
</ul>
<figure id="attachment_2695" aria-describedby="caption-attachment-2695" style="width: 831px" class="wp-caption aligncenter"><a href="https://read.transkribus.eu/wp-content/uploads/2019/03/Crowd-leert.png"><img decoding="async" class="wp-image-2695 size-full" src="https://read.transkribus.eu/wp-content/uploads/2019/03/Crowd-leert.png" alt="" width="831" height="234" /></a><figcaption id="caption-attachment-2695" class="wp-caption-text">Image credit: Amsterdam City Archives</figcaption></figure>
Amsterdam City Archives are interested in opening up access to the records of Amsterdam&#8217;s notaries, which span from the sixteenth to the twentieth century. These documents are ripe for further exploration for those interested in the rich social and economic history of the Dutch capital.  The ultimate aim is to create a fully searchable record of this precious handwritten collection.
The team have been working with our <a href="https://transkribus.eu/Transkribus/" target="_blank" rel="noopener noreferrer">Transkribus</a> platform to train HTR models to recognise different parts of this collection.
<ul>
<li><a href="https://twitter.com/voetnoot/status/1075340886698049537" target="_blank" rel="noopener noreferrer">Take a look at a recent example of a seventeenth century document recognised with a Character Error Rate of 6%. </a></li>
</ul>
The HTR models were used to generate automated transcripts of the documents. It is now up to volunteers to correct any errors made by the machine!
The project is hosted on VeleHanden, a successful crowdsourcing platform created by the company <a href="https://picturae.com/en/" target="_blank" rel="noopener noreferrer">Picturae</a>. Crowd leert computer lezen is directly connected to the Transkribus web interface, meaning that any changes made by volunteers can be fed straight back into the system to improve the automated recognition.
Anyone can take part in this new project and explore various difficulty levels to find documents they are interested in.  Volunteers collect points for their transcription work which can be redeemed at exhibitions and events at Amsterdam City Archives.
We are really looking forward to seeing what the computer can learn from the crowd&#8230;
Mark Ponte from Amsterdam City Archives gave us a sneak peak of the project at our recent Transkribus User Conference
<ul>
<li><a href="https://read.transkribus.eu/wp-content/uploads/2018/11/PONTE-AMSTERDAM.pdf" target="_blank" rel="noopener noreferrer">Check out conference slides from Amsterdam City Archives </a></li>
</ul>
<figure id="attachment_2696" aria-describedby="caption-attachment-2696" style="width: 1078px" class="wp-caption aligncenter"><a href="https://read.transkribus.eu/wp-content/uploads/2019/03/Amsterdam-City-Archives.png"><img decoding="async" class="wp-image-2696 size-full" src="https://read.transkribus.eu/wp-content/uploads/2019/03/Amsterdam-City-Archives.png" alt="" width="1078" height="670" /></a><figcaption id="caption-attachment-2696" class="wp-caption-text">CORNELIS STAAL 1749-1753 &#8211; 1 &#8211; Beginner &#8211; 13131 &#8211; A31239000579. Screenshot from Crowd leert computer lezen. Image credit: Amsterdam City Archives.</figcaption></figure>

+ Crowdsourcing with Transkribus at Amsterdam City Archives

Our latest milestone has put a big smile on our faces &#8211; there are now over 20,000 registered users of our <a href="https://transkribus.eu/Transkribus/" target="_blank" rel="noopener noreferrer">Transkribus</a> platform for Handwritten Text Recognition! People are working with Transkribus across the globe, using it to train hundreds of models to recognise texts of diverse dates, languages and styles.
Across the course of the READ project, we have welcomed over 13,000 new users of the platform and created a formidable network of people interested in opening up access to historical documents. We look forward to continued growth as we move into our next phase with <a href="https://read.transkribus.eu/2019/01/25/read-on-the-move-to-read-coop/" target="_blank" rel="noopener noreferrer">READ COOP</a>.
Become part of our community! Follow us on <a href="https://twitter.com/Transkribus" target="_blank" rel="noopener noreferrer">Twitter</a> or <a href="https://www.facebook.com/Transkribus-Platform-546945145805190/" target="_blank" rel="noopener noreferrer">Facebook</a> to keep up to date with our latest news.
<a href="https://read.transkribus.eu/wp-content/uploads/2019/03/digits-4014181_1920.jpg"><img decoding="async" class="aligncenter size-full wp-image-2699" src="https://read.transkribus.eu/wp-content/uploads/2019/03/digits-4014181_1920.jpg" alt="" width="1920" height="1280" /></a>

+ 20,000 tremendous Transkribus users!

The <a href="https://www.rcplondon.ac.uk/" target="_blank" rel="noopener noreferrer">Royal College of Physicians</a> has been devoted to advancing medicine for the past 500 years and has amassed outstanding historical collections of rare books, medical instruments and medicinal plant specimens.
The RCP has recently digitised the 6000 sheets from the (mostly) nineteenth-century Herbarium of the Pharmaceutical Society of Great Britain.  This collection comprises thousands of preserved plant specimens and their associated labels.
Dr Michael de Swiet, Dr Henry Oakley and Professor Anthony Dayan of the RCP then decided to work with the <a href="https://transkribus.eu/Transkribus/" target="_blank" rel="noopener noreferrer">Transkribus</a> team to try to recognise the text from the Herbarium collection.
The documents present various challenges for Handwritten Text Recognition (HTR) technology.  They contain a mix of printed and handwritten text (in ink and pencil), various languages, abbreviations and specialist vocabulary. They are also written in several (similar) hands.
A first HTR model was trained on 29,083 transcribed words  from the collection, using the pre-existing &#8216;English Writing M1’ model as part of the training process.  The <a href="http://blogs.ucl.ac.uk/transcribe-bentham/2018/02/23/project-update-bentham-vs-computer/" target="_blank" rel="noopener noreferrer">‘English Writing M1’ model is trained to recognise the writing of the English philosopher Jeremy Bentham (1748 – 1832) and his secretaries</a> – it is freely available to all Transkribus users for their experiments.
In the best cases, the resulting model can automatically transcribe pages from the collection with a Character Error Rate (CER) of around 10%.
<figure id="attachment_2566" aria-describedby="caption-attachment-2566" style="width: 629px" class="wp-caption aligncenter"><a href="https://read.transkribus.eu/wp-content/uploads/2019/01/RCP-Herbarium.png"><img decoding="async" class="wp-image-2566 size-full" src="https://read.transkribus.eu/wp-content/uploads/2019/01/RCP-Herbarium.png" alt="" width="629" height="627" /></a><figcaption id="caption-attachment-2566" class="wp-caption-text">Image from the Herbarium with an automated transcription of the label. Image credit: Royal College of Physicians.</figcaption></figure>
The team at the RCP are pleased with these results and would be happy if they could be shared and improved upon by other people working with Herbarium material.  If you would like to find out more about their work or have access to their HTR model, please contact the Transkribus team (<a href="mailto:email@transkribus.eu" target="_blank" rel="noopener noreferrer">email@transkribus.eu</a>).

+ Plant power! Results from the Royal College of Physicians&#8217; Herbarium

Another day, another news feature for Transkribus! <a href="https://www.ilfattoquotidiano.it/" target="_blank" rel="noopener noreferrer">il Fatto Quotidiano</a>, an Italian daily newspaper, recently published a summary of our huge achievements in Handwritten Text Recognition and invited readers to try out our technology.
<ul>
<li><a href="https://www.ilfattoquotidiano.it/2019/01/31/i-manoscritti-antichi-diventano-digitali-grazie-al-progetto-read-della-commissione-europea/4937668/" target="_blank" rel="noopener noreferrer">Read the article (in Italian)</a></li>
</ul>

+ Transkribus featured in Italian newspaper

After the tumult of World War One, the First Austrian Republic was declared in September 1919. Nearly 100 years later, the Austrian Parliamentary Directorate are making documents from this crucial period of national history available online.
They have recently digitised 9000 parliamentary papers from the First Republic and are working with our <a href="https://transkribus.eu/Transkribus/" target="_blank" rel="noopener noreferrer">Transkribus</a> platform to provide the fully-searchable text of this collection.
<figure id="attachment_2671" aria-describedby="caption-attachment-2671" style="width: 670px" class="wp-caption aligncenter"><a href="https://read.transkribus.eu/wp-content/uploads/2019/02/Austrian-Parliament.png"><img decoding="async" class="wp-image-2671 size-full" src="https://read.transkribus.eu/wp-content/uploads/2019/02/Austrian-Parliament.png" alt="" width="670" height="776" /></a><figcaption id="caption-attachment-2671" class="wp-caption-text">Document from the Meeting of the Provisional National Assembly on 6 February 1919. Image credit: L1.5 Kompetenzzentrum, Parlamentsdirektion.</figcaption></figure>
These parliamentary documents contain both handwritten and printed text (using the Fraktur font of Latin calligraphy).
5000 pages of the collection have now been processed in Transkribus with a mixture of Handwritten Text Recognition and our ABBYY FineReader OCR engine. 3,500 of these pages were then manually checked to ensure the accuracy of the recognised words.  Keywords have been captured and assigned to documents, allowing for further search options.  The next task is to record the titles and metadata of the documents to enhance their usefulness.
<ul>
<li><a href="https://www.parlament.gv.at/PAKT/DE/index.shtml?jsMode=&amp;xdocumentUri=&amp;filterJq=&amp;view=&amp;NRBR=NR&amp;GP=PN&amp;R_TYP=KRZL&amp;ITYP=ALLE&amp;INR=&amp;INRBIS=&amp;SUCH=&amp;listeId=120&amp;LISTE=Show+&amp;FBEZ=FP_020" target="_blank" rel="noopener noreferrer">Consult the documents on the Austrian Parliament website</a></li>
</ul>

+ Birth of the First Republic: Recognising Austrian Parliamentary papers

We’re sure many people know how challenging it can be to read historical manuscripts. Our new e-learning website, <a href="https://transkribus.eu/r/learn/" target="_blank" rel="noopener noreferrer">Transkribus LEARN</a>  is here to help! Transkribus LEARN does not replace systematic paleography training but it allows users to practice reading and transcribing individual words, learning as they go.
Learners can practice on a computer or mobile phone and check their learning progress.
Transkribus LEARN has two transcription modes – ‘Study’ and ‘Test’. In the former, users can guess and then reveal the transcription of individual words in a manuscript. In the latter, users will be prompted to transcribe the missing word in a series of examples. At the end, a score is received, along with a list of correct and incorrect answers. Users can keep studying and testing themselves, as often as they like.
<figure id="attachment_2666" aria-describedby="caption-attachment-2666" style="width: 600px" class="wp-caption aligncenter"><a href="https://read.transkribus.eu/wp-content/uploads/2019/02/Learn-Sutterlin.png"><img decoding="async" class="wp-image-2666" src="https://read.transkribus.eu/wp-content/uploads/2019/02/Learn-Sutterlin.png" alt="" width="600" height="567" /></a><figcaption id="caption-attachment-2666" class="wp-caption-text">Screenshot from Transkribus Learn. Sütterlin script from a 1926 Austrian recipe book.</figcaption></figure>
Levels of difficulty vary from simple, computer-generated cursive scripts to challenging historical documents in various languages and of different time periods. Users can also upload their own documents to the platform as a training exercise for students or volunteers.
To try it out:
<ul>
<li><a href="https://transkribus.eu/Transkribus/" target="_blank" rel="noopener noreferrer">Register for a free Transkribus account</a></li>
<li><a href="https://transkribus.eu/r/learn/" target="_blank" rel="noopener noreferrer">Login at Transkribus Learn (works on web or mobile)</a></li>
<li>Explore the site and choose a document to practice on</li>
</ul>
Happy learning one and all! We welcome any feedback at: <a href="mailto:learn@transkribus.eu" target="_blank" rel="noopener noreferrer">learn@transkribus.eu</a>

+ Learn how to read historical handwriting with Transkribus Learn

Our new <a href="https://www.facebook.com/Transkribus-Platform-546945145805190/" target="_blank" rel="noopener noreferrer">Facebook page</a> can help you keep up-to-date with the latest goings-on with Transkribus and the READ project. Like and follow us to help us grow our community!
<a href="https://read.transkribus.eu/wp-content/uploads/2019/02/Facebook.png"><img decoding="async" class="aligncenter size-full wp-image-2657" src="https://read.transkribus.eu/wp-content/uploads/2019/02/Facebook.png" alt="" width="734" height="295" /></a>
You can also discuss Transkribus tips and tricks in the user-led Facebook group,<a href="https://www.facebook.com/groups/614090738935143//" target="_blank" rel="noopener noreferrer"> Transkribus Users</a>.

+ Like us on Facebook!

At the READ project, we believe in using cutting-edge technology to help people study a rich variety of historical documents.
The <a href="https://cvl.tuwien.ac.at/" target="_blank" rel="noopener noreferrer">Computer Vision Lab</a> at the Technical University of Vienna (one of the READ project partners) have developed the <a href="https://scantent.eu/en/#docscan" target="_blank" rel="noopener noreferrer">DocScan</a> mobile app for this very purpose.
The app automatically detects the page area of a document in milliseconds and provides real-time feedback on the quality of the image. It also has an auto-shoot feature which will take a picture every time a page is turned. It works especially well when used alongside our <a href="https://scantent.eu/" target="_blank" rel="noopener noreferrer">ScanTent</a> device (also developed by the Computer Vision Lab), which holds a mobile phone in place above a historical document and allows for hands-free scanning.
The remarkable potential of these tools was revealed at our recent <a href="https://read.transkribus.eu/transkribus-user-conference-2018/" target="_blank" rel="noopener noreferrer">Transkribus User Conference</a>. Dirk Alvermann from the <a href="https://ub.uni-greifswald.de/en/" target="_blank" rel="noopener noreferrer">University of Greifswald Library</a> (one of <a href="https://read.transkribus.eu/network/" target="_blank" rel="noopener noreferrer">READ&#8217;s MOU partners</a>) presented the results of an experiment his team had been conducting.
The idea is that archival users can work with DocScan and the ScanTent to digitise historical documents with their mobile phone and then share the resulting images directly with the archive.
As shown in the below video, Greifswald University Library assigned a QR code to a set of documents and asked users to scan this code with the DocScan app before they started digitising documents. Once images were scanned using DocScan and the ScanTent, they were uploaded to our <a href="https://transkribus.eu/Transkribus/" target="_blank" rel="noopener noreferrer">Transkribus</a> platform and became available to view and transcribe in the Transkribus Web interface. The library was then able to create links in its digital repository, connecting archival metadata with the digitised images on Transkribus Web. A future version of DocScan will make it easier for images to be ingested directly into archival systems.
Dirk Alvermann emphasised that this workflow could be incredibly beneficial for small archives who lack funding for digitisation.  Whilst user-generated content is not a substitute for a full digitisation strategy, it has the advantage of creating new resources and engaging interested archival users.
The DocScan app is available to download now, free of charge. The ScanTent is still in development and will be available for sale and hire later in 2019.
<iframe title="DocScan Greifswald" width="800" height="450" src="https://www.youtube.com/embed/BVnKnsWUHOM?feature=oembed" frameborder="0" allow="accelerometer; autoplay; clipboard-write; encrypted-media; gyroscope; picture-in-picture" allowfullscreen></iframe>
<ul>
<li><a href="https://read.transkribus.eu/wp-content/uploads/2018/11/ALVERMANN-GREIFSWALD.pdf" target="_blank" rel="noopener noreferrer">Read Dirk Alvermann&#8217;s slides from the Transkribus User Conference</a></li>
<li><a href="https://play.google.com/store/apps/details?id=at.ac.tuwien.caa.docscan&amp;hl=en&amp;pcampaignid=MKT-Other-global-all-co-prtnr-py-PartBadge-Mar2515-1" target="_blank" rel="noopener noreferrer">Download the DocScan app for free (Android only)</a></li>
<li><a href="https://scantent.eu/en/" target="_blank" rel="noopener noreferrer">Read more about the ScanTent</a></li>
</ul>

+ Crowdsourced digitisation with DocScan

Our <a href="https://transkribus.eu/Transkribus/" target="_blank" rel="noopener noreferrer">Transkribus</a> platform for Handwritten Text Recognition (HTR) is used by thousands of researchers and archivists all over the world. And we&#8217;ve just been featured on the news on the New Zealand television network TVNZ.
<iframe src="//players.brightcove.net/963482464001/m0mCB4JBd_default/index.html?videoId=5995249415001" frameborder="0" allowfullscreen="allowfullscreen"></iframe>
<a href="http://archives.govt.nz/" target="_blank" rel="noopener noreferrer">Archives New Zealand</a> explained how they have been experimenting with Transkribus to recognise different historical collections. They underlined some of the huge benefits of working with Transkribus &#8211; transcription is sped up, physical documents receive less wear and tear and most importantly, historical treasures become much easier to access!
<ul>
<li><a href="https://www.tvnz.co.nz/one-news/new-zealand/new-technology-helps-decipher-historic-hand-written-documents-national-archive?variant=tb_v_3" target="_blank" rel="noopener noreferrer">Read more in the 1 News Now article</a></li>
</ul>

+ Transkribus around the globe &#8211; coverage on TVNZ

With thousands of <a href="https://transkribus.eu/Transkribus/" target="_blank" rel="noopener noreferrer">Transkribus</a> users working all over the world, there is huge potential for collaborative work on the automated recognition of historical documents.
<a href="https://twitter.com/thist_uzh" target="_blank" rel="noopener noreferrer">Dr Tobias Hodel</a> (State Archives of Zurich, University of Zurich) has set up the &#8216;Gothic Hands&#8217; working group with this mind, hoping to improve recognition of medieval Gothic script. The &#8216;Comb_Gothic_Bookwriting&#8217; model has been trained on different sets of medieval scripts and is already available to all Transkribus users. In the best cases, it can produce automated transcripts with a Character Error Rate of less than 10%.
We are looking for more users to join this working group and share images and transcripts of Gothic scripts written between the 11th and 15th century. The current model has been trained primarily on German language material, so we are especially keen to receive documents written in Latin.  
The latest contributor to the working group is <a href="http://achilleid.unige.ch/welcome" target="_blank" rel="noopener noreferrer">Digital Statius: The Achilleid,</a> a project which is producing a digital edition of Achilleid, an unfinished epic poem written in Latin, in which the poet Statius (later 1st c. AD), narrates the childhood of Achilles and the stay of the hero on the island of Scyros. This text was part of the school curriculum in the Middle Ages, before losing its status as a classic. The project, funded by the <a href="http://p3.snf.ch/project-170010" target="_blank" rel="noopener noreferrer">Swiss National Science Foundation (SNSF)</a> and based at the University of Geneva, aims to produce a new critical edition of the Achilleid, fully and exclusively digital, which takes into account the complete manuscript tradition of the poem (224 manuscripts, c. 8000 images). The open access digital critical edition will include a new text, a full interactive apparatus criticus, comparative visualization of numerous readings, comments, translations, links to other tools and/or platforms, and the images of the largest possible number of manuscripts.
If you work with Gothic script, you can join the team behind the Achilleid edition and many others by becoming part of the &#8216;Gothic Hands&#8217; working group.
To participate in the group, you can:
<ul>
<li>share existing training data that you have already prepared in Transkribus</li>
<li>prepare new images and transcripts in Transkribus in the &#8216;Gothic Hands&#8217; collection</li>
<li>send over files containing images and transcripts which can be matched automatically and converted into training data</li>
</ul>
Please contact Tobias Hodel (<a href="mailto:tobias.hodel@hist.uzh.ch" target="_blank" rel="noopener noreferrer">tobias.hodel@hist.uzh.ch</a>) with any questions about the group.
Working together gives us a great chance to transcribe and search medieval documents more efficiently!
<a href="https://read.transkribus.eu/wp-content/uploads/2019/01/Digital-Statius.png"><img decoding="async" class="aligncenter size-full wp-image-2626" src="https://read.transkribus.eu/wp-content/uploads/2019/01/Digital-Statius.png" alt="" width="657" height="559" /></a>

+ Medievalists! Share data with our working group to improve Handwritten Text Recognition

The READ consortium together with several other institutions is currently preparing the foundation of a legal entity (working title: READ-COOP) which will serve as the basis for sustaining and further developing the Transkribus platform and related services.
The governance model will be based on the EU directive for European Cooperative Societies (SCE). Though the SCE will be set up according to EU law it will be open to members outside of the European Community as well.
An SCE
<ul>
<li>is a legal entity that allows its members to carry out common activities, while preserving their independence</li>
<li>has the principal objective of satisfying its members&#8217; needs and not the return of capital investment</li>
<li>allows members to benefit proportionally to their profit and not to their capital contribution.</li>
</ul>
<a href="https://read.transkribus.eu/coop/">Read more</a>

+ READ on the move to READ-COOP

<a href="https://www.newseye.eu/consortium/university-of-innsbruck/#c82" target="_blank" rel="noopener noreferrer">Günter Mühlberger</a>, coordinator of READ and head of the Digital Humanities Research Center at the University of Innsbruck has recently been interviewed on a new podcast (in German).
The interview was recorded by the <a href="https://www.newseye.eu/" target="_blank" rel="noopener noreferrer">NewsEye</a> project which like READ, is funded by European Union’s Horizon 2020 scheme. NewsEye aims to use digital tools to provide enhanced access to digitised historical newspapers and the project will build upon READ&#8217;s existing achievements in relating to the automated recognition of printed text.
<ul>
<li><a href="https://www.newseye.eu/media/" target="_blank" rel="noopener noreferrer">Listen to the podcast (in German).</a></li>
</ul>

+ Podcast with READ project coordinator

We have another new How to Guide for users of our <a href="https://transkribus.eu/Transkribus/" target="_blank" rel="noopener noreferrer">Transkribus</a> platform.  This time we’re showing you how to enrich documents with structural tags like ‘paragraph’, ‘heading’, or ‘footer’.
<ul>
<li><a href="https://transkribus.eu/wiki/images/f/f7/How_to_Use_the_Structural_Tagging_Feature.pdf" target="_blank" rel="noopener noreferrer">How to Use the Structural Tagging Feature</a></li>
</ul>
In the near future, it will be possible to train models to automatically recognise the structure of historical documents.  Adding structural tags creates training data for this process.  If you work with this feature, there is no need to tag every element of your documents – just focus on marking up the sections that are of interest to you.
If you have any questions about structural tags, the Transkribus team are here to help (<a href="mailto:email@transkribus.eu">email@transkribus.eu</a>)

+ Learn how to add structural tags to documents in Transkribus

Prototypes of the <a href="https://scantent.eu/en/" target="_blank" rel="noopener noreferrer">ScanTent</a>, our device for digitising documents with a mobile phone, have been popping up all over Europe over the past year. And in December 2018, the first ScanTent made it to Africa!
<a href="https://www.kcl.ac.uk/artshums/depts/history/people/staff/academic/hiribarrenv/hiribarrenv.aspx" target="_blank" rel="noopener noreferrer">Dr Vincent Hiribarren (King&#8217;s College London)</a> took the tent to the town of Kita in western Mali to try it out before using the professional equipment (cameras, tripod, scanner) provided by the <a href="https://eap.bl.uk/" target="_blank" rel="noopener noreferrer">Endangered Archives Programme</a> project called &#8216;Recovering the rich local history of Kita (Mali) through the salvaging of its archival heritage&#8217;.  This grant is held and directed by <a href="https://www.soas.ac.uk/staff/staff69183.php" target="_blank" rel="noopener noreferrer">Dr Marie Rodet (SOAS, University of London)</a>.
The Endangered Archives Programme at <a href="https://www.bl.uk/" target="_blank" rel="noopener noreferrer">The British Library</a> awards annual grants to preserve archival material that is at risk of destruction or neglect. This funding means that endangered archival collections can be transferred to new homes, digitised and deposited at local institutions and in The British Library.
The ScanTent is a portable piece of equipment which holds a user&#8217;s phone in place above a historical document, providing a consistent source of light and leaving users with their hands free to turn pages or move documents around.  The advantages of the ScanTent become even greater when it is used in conjunction with our <a href="https://scantent.eu/en/#docscan" target="_blank" rel="noopener noreferrer">DocScan</a> mobile app.  DocScan automatically detects the page area of a document and provides real-time feedback on image quality.  It also has an auto-shoot feature which will take a photo every time a page is turned.  <a href="https://transkribus.eu/Transkribus/" target="_blank" rel="noopener noreferrer">Transkribus</a> users can upload images to the platform directly from DocScan and these images can then be used for <a href="https://transkribus.eu/wiki/images/5/50/How_To_Transcribe_Documents_with_Transkribus.pdf" target="_blank" rel="noopener noreferrer">training an Automated Text Recognition model.</a>
Dr Hiribarren installed DocScan on his phone in advance of his trip and was then able to set up the ScanTent quickly on location in Kita and start scanning! This experiment really shows that these tools have huge potential to open up access to unique collections of historical material all over the globe.
DocScan is <a href="https://play.google.com/store/apps/details?id=at.ac.tuwien.caa.docscan&amp;hl=en&amp;pcampaignid=MKT-Other-global-all-co-prtnr-py-PartBadge-Mar2515-1" target="_blank" rel="noopener noreferrer">available now</a>, free of charge.  The ScanTent is still in development and units will be available for sale and hire later in 2019.
<figure id="attachment_2532" aria-describedby="caption-attachment-2532" style="width: 1200px" class="wp-caption aligncenter"><a href="https://read.transkribus.eu/wp-content/uploads/2019/01/IMG-20190110-WA0001.jpg"><img decoding="async" class="size-full wp-image-2532" src="https://read.transkribus.eu/wp-content/uploads/2019/01/IMG-20190110-WA0001.jpg" alt="" width="1200" height="1600" /></a><figcaption id="caption-attachment-2532" class="wp-caption-text">Testing out the ScanTent in Kita (Mali). Image credit: Vincent Hiribarren.</figcaption></figure>
Find out more:
<ul>
<li><a href="https://scantent.eu/en/" target="_blank" rel="noopener noreferrer">ScanTent and DocScan website</a></li>
<li><a href="https://play.google.com/store/apps/details?id=at.ac.tuwien.caa.docscan&amp;hl=en&amp;pcampaignid=MKT-Other-global-all-co-prtnr-py-PartBadge-Mar2515-1" target="_blank" rel="noopener noreferrer">Download DocScan app (Android only)</a></li>
<li><a href="https://eap.bl.uk/" target="_blank" rel="noopener noreferrer">Endangered Archives Programme</a></li>
</ul>

+ ScanTent makes it to Mali, West Africa!

We&#8217;re celebrating the New Year with the release of several video tutorials designed to help new users navigate our <a href="https://transkribus.eu/Transkribus/" target="_blank" rel="noopener noreferrer">Transkribus</a> platform.
If you have a few minutes, you can get a nice overview of stages needed to automatically recognise and search handwritten and printed historical texts in Transkribus.
How to use Transkribus in 10 steps
<iframe title="How To Use Transkribus in 10 steps (updated!)" width="800" height="450" src="https://www.youtube.com/embed/8Ei0a7WIlTI?feature=oembed" frameborder="0" allow="accelerometer; autoplay; clipboard-write; encrypted-media; gyroscope; picture-in-picture" allowfullscreen></iframe>
Segmentation
<iframe title="Segmentation Tutorial Transkribus" width="800" height="450" src="https://www.youtube.com/embed/b0MqNwSIKC0?feature=oembed" frameborder="0" allow="accelerometer; autoplay; clipboard-write; encrypted-media; gyroscope; picture-in-picture" allowfullscreen></iframe>
Training a Handwritten Text Recognition model
<iframe title="Model Training Tutorial Transkribus" width="800" height="600" src="https://www.youtube.com/embed/21XMn9JoDEM?feature=oembed" frameborder="0" allow="accelerometer; autoplay; clipboard-write; encrypted-media; gyroscope; picture-in-picture" allowfullscreen></iframe>
Using Handwritten Text Recognition models
<iframe title="Using Handwritten Text Recognition Models for your documents - Tutorial Transkribus" width="800" height="600" src="https://www.youtube.com/embed/XWdj81Nfwtw?feature=oembed" frameborder="0" allow="accelerometer; autoplay; clipboard-write; encrypted-media; gyroscope; picture-in-picture" allowfullscreen></iframe>
Keyword Spotting
<iframe title="Keyword Spotting Tutorial Transkribus" width="800" height="600" src="https://www.youtube.com/embed/3RLYgZy0Pwc?feature=oembed" frameborder="0" allow="accelerometer; autoplay; clipboard-write; encrypted-media; gyroscope; picture-in-picture" allowfullscreen></iframe>
If you need extra help with Transkribus, please check out our detailed <a href="https://transkribus.eu/wiki/index.php/How_to_Guides" target="_blank" rel="noopener noreferrer">How to Guides.</a>

+ Latest Transkribus video tutorials

What&#8217;s your name?
Johanna Walcher
<a href="https://read.transkribus.eu/wp-content/uploads/2018/07/1908227_747693485248434_1382826270_n.jpg"><img decoding="async" class="size-medium wp-image-2080 alignleft" src="https://read.transkribus.eu/wp-content/uploads/2018/07/1908227_747693485248434_1382826270_n-264x300.jpg" alt="" width="264" height="300" /></a>
&nbsp;
&nbsp;
&nbsp;
&nbsp;
&nbsp;
&nbsp;
&nbsp;
&nbsp;
&nbsp;
&nbsp;
Where do you work?
The University of Innsbruck.
Tell us a bit about your background…
I did a bachelor&#8217;s degree in Transcultural Communication for Italian and Russian Language. Currently I am working towards my master’s degree in Media Studies and also study Philosophy. In my leisure time I love doing sports, preferably in the mountains. Hiking, running, mountain biking, skiing and yoga are great! But I can also sit still and spend a whole day reading books and newspapers. Whenever I can I travel the world and spend as much time as I can with my friends and family.
What is your role in the READ project?
In the READ project I am responsible for the <a href="https://transkribus.eu/wiki/index.php/How_to_Guides" target="_blank" rel="noopener noreferrer">Transkribus How to Guides,</a> which should make life easier for users. I also run Transkribus <a href="https://youtu.be/sHRiluoku4A" target="_blank" rel="noopener noreferrer">webinars</a> where I present the READ project and show interested people how to use the platform. As a member of the Dissemination Working Group, I am involved in spreading the word about the READ project.
What is at the top of your to-do list at the moment?
My current project is recording short screencast videos, which explain the different steps of the Transkribus workflow.  The first ones are already up on our <a href="https://www.youtube.com/channel/UC-txVgM31rDTGlBnH-zpPjA" target="_blank" rel="noopener noreferrer">YouTube channel</a> soon.
What do you like best about working on READ?
All the interesting projects around READ and the inspiring people I work with.
If you could do another job for just one day, what would it be?
Good question! There are too many jobs in the world I would love to try and it would take me too much time to choose.  Working hours are expensive, so I will rather save READ some money and do my actual job, which I like a lot! 🙂
What can you see out of the window of your office? 
<a href="https://read.transkribus.eu/wp-content/uploads/2018/07/View-Innsbruck-Johanna.jpg"><img decoding="async" class="wp-image-2081 size-large alignleft" src="https://read.transkribus.eu/wp-content/uploads/2018/07/View-Innsbruck-Johanna-1024x768.jpg" alt="" width="720" height="540" /></a>
&nbsp;
&nbsp;
&nbsp;
&nbsp;
&nbsp;
&nbsp;
&nbsp;
&nbsp;
&nbsp;
&nbsp;
&nbsp;
&nbsp;
&nbsp;
&nbsp;
&nbsp;
&nbsp;
&nbsp;
Thanks Johanna!

+ Meet the READ project partners Johanna Walcher

At the READ project we are committed to sharing data and working collaboratively to improve the recognition of handwritten historical documents.
With this in mind, two of the project’ research groups have uploaded data relating to recent computer science competitions in handwriting and document layout analysis.
Data from the Pattern Recognition and Human Language Technology group at the Universitat Politècnica de València and the Computational Intelligence Lab (CITLab) at the University of Rostock:
<ul>
<li>International Conference on Frontiers in Handwriting Recognition (ICFHR) 2014 <a href="http://doi.org/10.5281/zenodo.44519">http://doi.org/10.5281/zenodo.44519</a></li>
<li>International Conference on Document Analysis and Recognition (ICDAR) 2015 <a href="http://doi.org/10.5281/zenodo.248733">http://doi.org/10.5281/zenodo.248733</a></li>
<li>International Conference on Frontiers in Handwriting Recognition (ICFHR) 2016 <a href="http://doi.org/10.5281/zenodo.1164045">http://doi.org/10.5281/zenodo.1164045</a></li>
<li>International Conference on Document Analysis and Recognition (ICDAR) 2017 <a href="http://doi.org/10.5281/zenodo.835489">http://doi.org/10.5281/zenodo.835489</a></li>
<li>International Conference on Frontiers in Handwriting Recognition (ICFHR) 2018 <a href="https://doi.org/10.5281/zenodo.1442182">https://doi.org/10.5281/zenodo.1442182</a></li>
</ul>
Check out the <a href="https://zenodo.org/communities/scriptnet/?page=1&amp;size=20">ScriptNet-READ community on Zenodo</a> for more of the data that READ project researchers are using for their experiments.

+Sharing data for Handwritten Text Recognition

Segmentation is a crucial stage of working with Handwritten Text Recognition in our <a href="https://transkribus.eu/Transkribus/" target="_blank" rel="noopener noreferrer">Transkribus</a> platform.  Digitised images of historical documents must be segmented into text regions, lines and baselines before they can be transcribed manually or automatically.  Segmentation can be performed automatically by the software to a high level of accuracy.  For more complex documents, users may then need to make some manual corrections &#8211; moving or deleting baselines for example.
If you&#8217;re new to segmentation in Transkribus, we have a new video tutorial which will help you get started.
You can find out more about working with Transkribus in our <a href="https://transkribus.eu/wiki/index.php/Main_Page" target="_blank" rel="noopener noreferrer">How to Guides</a>.
<iframe title="Segmentation Tutorial Transkribus" width="800" height="450" src="https://www.youtube.com/embed/b0MqNwSIKC0?feature=oembed" frameborder="0" allow="accelerometer; autoplay; clipboard-write; encrypted-media; gyroscope; picture-in-picture" allowfullscreen></iframe>

+ A new video tutorial helps with segmentation in Transkribus

Yes, you read that correctly &#8211; our <a href="https://transkribus.eu/Transkribus/" target="_blank" rel="noopener noreferrer">Transkribus</a> platform can indeed recognise printed Indian texts.
Conventional OCR software usually struggles to decipher the complexities of South Asian scripts.  Two projects have recently been working with nineteenth-century printed texts in Transkribus with the hope of getting better results.  Using images and transcripts from a collection, Transkribus users can train a model to recognise printed text of any type.
First of all, The British Library&#8217;s <a href="https://www.bl.uk/early-indian-printed-books" target="_blank" rel="noopener noreferrer">Two Centuries of Indian Print</a> project is creating a digitised collection of works published in South Asia in the eighteenth and nineteenth centuries.  The project team trained a text recognition model in Transkribus with 50 pages (containing 5,700 words) of digitised images and transcripts from Bengali books.  The resulting model can produce transcripts of page from the collection with an average Character Error Rate of 21%.  Although this is a relatively high error rate, the team are planning to retrain the model by creating more pages of training data and focusing on improving the recognition of elements of the Bengali characters which were sometimes missed by the software.
The Naval Kishore Press was a nineteenth-century publishing house which brought works on various subjects to market in Hindi, Urdu, Arabic, Persian and Sanskrit. Part of its output are held by the library of the South Asia Institute (SAI) at Heidelberg University.  The South Asia Institute library and Heidelberg University Library are collaborating on the <a href="https://digi.ub.uni-heidelberg.de/en/sammlungen/suedasien/navalkishore.html" target="_blank" rel="noopener noreferrer">Naval Kishore Press – digital</a> project, working to produce digitised and machine-readable text for a selection of texts published by this press.  The project team used 200 pages of images and transcripts to train a model in Transkribus to recognise Hindi and Sanskrit text.  This model can produce transcripts of the collection with a Character Error Rate of around 5%.  Fully searchable images and transcripts from the collection are now available to consult, download and annotate on Heidelberg University library&#8217;s online catalogue.
Read more:
<ul>
<li><a href="https://blogs.bl.uk/digital-scholarship/2018/10/using-transkribus-for-automated-text-recognition-of-historical-bengali-books.html">Bengali books at The British Library &#8211; blog post by Tom Derrick, Digital Curator at The British Library</a></li>
<li><a href="https://iias.asia/the-newsletter/article/naval-kishore-press-digital-hidden-treasure-open-access" target="_blank" rel="noopener noreferrer">Naval Kishore Press digital project &#8211; article by Nicole Merkel-Hilf in the newsletter of the International Institute for Asian Studies</a></li>
</ul>

+ Recognising printed Asian texts with Transkribus

Many new users are registering for a <a href="https://transkribus.eu/Transkribus/" target="_blank" rel="noopener noreferrer">Transkribus</a> account every day and our How to Guides are there to help everyone get to grips with Handwritten Text Recognition technology for historical documents.
All of our How to Guides are now available in English and German.
Our introductory guide, ‘How to use Transkribus in 10 steps’ is also available in French.
You can find all of our How to Guides on the Transkribus wiki.
<ul>
<li><a href="https://transkribus.eu/wiki/index.php/How_to_Guides" target="_blank" rel="noopener noreferrer">How to Guides </a></li>
</ul>
Our thanks go to Régis Schlagdenhaffen for the French translation.

+Transkribus How to Guides now available in German (and French)

The READ project is a big proponent of digitisation on demand using smartphones.
A typical mobile phone camera can capture relatively high-quality images of historical documents, which can then be used for preservation, research and even as training data for Automated Text Recognition using our <a href="https://transkribus.eu/Transkribus/" target="_blank" rel="noopener noreferrer">Transkribus</a> platform.
<a href="https://cvl.tuwien.ac.at/" target="_blank" rel="noopener noreferrer">The Computer Vision Lab</a> at the Technical University of Vienna (one of the READ project partners) have created the <a href="http://scantent.eu/" target="_blank" rel="noopener noreferrer">ScanTent</a> device and the <a href="https://scantent.eu/en/#docscan" target="_blank" rel="noopener noreferrer">DocScan mobile app</a> to make it easier for people to digitise documents in this way.
<figure id="attachment_1966" aria-describedby="caption-attachment-1966" style="width: 300px" class="wp-caption aligncenter"><a href="https://read.transkribus.eu/wp-content/uploads/2018/06/20180531_161544.jpg"><img decoding="async" class="size-medium wp-image-1966" src="https://read.transkribus.eu/wp-content/uploads/2018/06/20180531_161544-300x169.jpg" alt="" width="300" height="169" /></a><figcaption id="caption-attachment-1966" class="wp-caption-text">The ScanTents</figcaption></figure>
We were happy to receive a positive enquiry about these tools, highlighting their potential to capture unique records that might otherwise be lost.
Stefan Krüger from Germany got in touch after he had digitised his grandfather&#8217;s dissertation using his mobile phone and used Transkribus to recognise the text with OCR.  Herbert Rechner completed his dissertation in 1927 just before the rise of the Nazis, on the radical topic of the ‘the sexual causes of offences’.  Although Stefan was never able to meet his grandfather, he is interested in researching his history and is hopeful that Transkribus might be able to help recognise personal handwritten papers one day.
<a href="https://read.transkribus.eu/wp-content/uploads/2018/11/JG1_1654T1-18.jpg"><img decoding="async" class="wp-image-2446 alignleft" src="https://read.transkribus.eu/wp-content/uploads/2018/11/JG1_1654T1-18.jpg" alt="" width="167" height="250" /></a>
Stefan wrote&#8230;
&#8216;After a long search I found the 90 years old dissertation of my grandfather in the German National Library in Leipzig and (in bowing to the performance of my ancestor) digitally reproduced the work. The Transkribus project helped me a lot with its outstanding recognition rate.
I photographed the booklet (about 100 pages) freehand with glass plate and smartphone (CamScanner) and re-set it in InDesign after text recognition.
With this work it became clear to me that we are experiencing a scientific break: everything that is not digitally available in scientific literature will disappear in the cognition-sinking. It is simply no longer taken into account in the scientific knowledge and research process. In the case of topics relating to electronics, space travel and other &#8220;more modern&#8221; developments, this may be easy to accept.
With all historically relevant things, however, this is painful.
That&#8217;s why I find your low-level effort with high-tech solutions very interesting. I would like to test your tent and the app. My thought is that actually (at least) everyone who has enjoyed an academic education should participate in the digital processing of his work and other literature. If you could make such a crowd thing out of it, then a big stock of literature could actually be worked on. So I am happy to participate in your developments in this sense.
With cordial greetings
Stefan Krüger&#8217;
Translated from German with <a href="http://www.DeepL.com/Translator">www.DeepL.com/Translator</a>
Thank you to Stefan for this feedback, which shows how Transkribus can help individuals to digitise and recognise exceptional historical documents.
<figure id="attachment_2447" aria-describedby="caption-attachment-2447" style="width: 668px" class="wp-caption aligncenter"><a href="https://read.transkribus.eu/wp-content/uploads/2018/11/Rechner_Herbert.png"><img decoding="async" class="wp-image-2447" src="https://read.transkribus.eu/wp-content/uploads/2018/11/Rechner_Herbert.png" alt="" width="668" height="1000" /></a><figcaption id="caption-attachment-2447" class="wp-caption-text">A page from Herbert Rechner&#8217;s dissertation, digitised with a smartphone. Image credit: Stefan Krüger</figcaption></figure>
If you would like to try digitising documents with a mobile phone, the DocScan app is available to download now free of charge (Android only). The ScanTent is still in development and units should be available for sale over the next few months.
Find out more:
<ul>
<li><a href="https://scantent.eu/en/" target="_blank" rel="noopener noreferrer">Consult the ScanTent website</a></li>
<li><a href="https://play.google.com/store/apps/details?id=at.ac.tuwien.caa.docscan&amp;hl=en&amp;pcampaignid=MKT-Other-global-all-co-prtnr-py-PartBadge-Mar2515-1" target="_blank" rel="noopener noreferrer">Download the DocScan app for free (Android only)</a></li>
<li><a href="https://read.transkribus.eu/2018/06/13/report-from-2018-scanathon/" target="_blank" rel="noopener noreferrer">Read a report on the 2018 Scanathon where we tested these tools.</a></li>
</ul>

+ Preserving our cultural heritage with a smartphone

In sixteenth- and seventeenth-century Spain, there was a significant surge of thousands of theatrical productions. This period has become known as the Spanish Golden Age.  Thanks to a new protoype web tool, anyone can now search through 40,000 images from a significant digitised collection of manuscripts relating to this period of Spanish history.  This tool uses cutting-edge Keyword Spotting technology, allowing users to search images which have  never before been transcribed.
<ul>
<li><a href="http://prhlt-carabela.prhlt.upv.es/tso/" target="_blank" rel="noopener noreferrer">Start searching</a></li>
</ul>
This tool is a collaboration between the <a href="https://www.prhlt.upv.es/wp/" target="_blank" rel="noopener noreferrer">Pattern Recognition and Human Language Technology research centre</a> at the Universitat Politecnica de Valencia (one of the <a href="https://read.transkribus.eu/network/" target="_blank" rel="noopener noreferrer">READ partners</a>), the <a href="http://www.bne.es/en/Inicio/" target="_blank" rel="noopener noreferrer">National Library of Spain</a> and the <a href="http://prolope.uab.cat/" target="_blank" rel="noopener noreferrer">PROLOPE</a> research group (both <a href="https://read.transkribus.eu/network/" target="_blank" rel="noopener noreferrer">READ MOU partners</a>).
The PRHLT research centre has treated these manuscripts with advanced text recognition and probabilistic word indexing technology.  This sophisticated form of searching is often called Keyword Spotting. It is more powerful than a conventional full-text search because it uses statistical models trained for text recognition to search through probability values assigned to character sequences (words), considering most possible readings of each word on a page.
<figure id="attachment_2396" aria-describedby="caption-attachment-2396" style="width: 876px" class="wp-caption aligncenter"><a href="https://read.transkribus.eu/wp-content/uploads/2018/11/KWS-Madrid.png"><img decoding="async" class="wp-image-2396 size-full" src="https://read.transkribus.eu/wp-content/uploads/2018/11/KWS-Madrid.png" alt="" width="876" height="475" /></a><figcaption id="caption-attachment-2396" class="wp-caption-text">Keyword Spotting for the word &#8216;Madrid&#8217;.</figcaption></figure>
The 40,000 pages currently available for searching represents about half of the collection.  More documents from the collection will be processed in this way if further funding can be found.
The release of this Keyword Spotting tool coincides with a <a href="http://www.bne.es/es/Actividades/Exposiciones/Exposiciones/Exposiciones2018/lope-teatro-siglo-de-oro.html" target="_blank" rel="noopener noreferrer">new exhibition at the National Library of Spain</a> all about the Spanish Golden Age which runs until March 2019.  The exhibition will combine original manuscripts with digital displays.  The PRHLT team have a created an online quiz (in Spanish) for the exhibition which asks users to work with the Keyword Spotting too to find out which words appear frequently or in combination.
<ul>
<li><a href="http://prhlt-carabela.prhlt.upv.es/quizz-tso" target="_blank" rel="noopener noreferrer">Try out the quiz.</a></li>
</ul>
If you are interested in Keyword Spotting, check out other tools constructed by the PRHLT team relating to:
<ul>
<li><a href="http://prhlt-carabela.prhlt.upv.es/himanis/" target="_blank" rel="noopener noreferrer">Medieval chancery records</a></li>
<li><a href="https://read.transkribus.eu/2018/10/15/searching-benthams-manuscripts-kws/" target="_blank" rel="noopener noreferrer">Writings of English philosopher Jeremy Bentham.</a></li>
</ul>

Transkribus Blog

Start unlocking the past with Transkribus