Michalcová, Anna; Spěváčková, Martina; Kramarić, Martina; Mikloš, Jozef; Mizerová, Julie; Katscher, Daniel; Filipová, Jitka; Pasáčková, Eva; Vassmar, Martina; Pěnkavová, Eliška; Glaser, Leon; Vučičová, Klára · PyLaia · Published January 25, 2024
Old Czech Handwriting (with spaces), 2nd version
Text Recognition
Description
This model is based on ground truth from three Old Czech manuscripts:
1) Padeřov Bible (Vienna, Austrian National Library, shelfmark Cod. 1175, 1432–1435)
2) Cistercian Bible (New York, The Morgan Library & Museum, shelfmark MS M.752, 1456),
3) New Testament of Martin Lupáč (Vienna, Austrian National Library, shelfmark Cod. 1175, 1440).
It can be used on Old Czech manuscripts written in bastard, primarily from the 14th and 15th centuries.
The transcription rules were based on semi-diplomatic transcription rules set by Pero OCR and “Směrnice pro vydávání starších českých textů” by Jiří Daňhelka. https://vokabular.ujc.cas.cz/moduly/edicnipoznamka.aspx?id=DanhelkaSmernice
Abbreviations were tagged and expanded. This model was trained on transcription, where spaces were created between words based on Jiří Daňhelka’s semi-diplomatic transcription rules (see above), regardless of the form in the manuscripts. Produced within the HTR Winter School 2023.
Try this model
Use this modelOpen in Transkribus
Low error rate5.1% CER
Character Error Rate (CER) measures the percentage of characters incorrectly recognised. Lower is better. This model scored 5.1% on its validation set. As a rule of thumb, a CER below 10% is considered good for most handwritten material.
Measured on the model's own validation data. Results on your documents may differ depending on handwriting style, document condition, language, and how closely your material resembles the training data.
Words97,997
Lines18,656
Training Pages194
Model ID58856