Creator(s)
Newseye-project
Language(s)
French
Centuries
18th, 19th, 20th
CER on Validation Set
4.2%
Size (Nr. of Words)
434,100
Model ID
37747
The model works well with French print from late 18th century to mid of 20th century. For standard text in French newspapers from that time error rates much below 1% were measured.
The model was created in the NewsEye project and is based on training data coming from the digital library Gallica of the French National Library (BnF). 434100 words have been trained and the CER on the validation set is 4.20%.
You can use this model to automatically transcribe Handwritten documents with Handwritten Text Recgnition in Transkribus.