Skip to content
  • Pricing
vadimverenich · PyLaia · Published November 17, 2026

Cyrillic handwriting_14th-17th_C

Text Recognition

Description

Handwritten text recognition model (PyLaia) for Ruthenian (Old Belarusian / Old Ukrainian) chancery cursive (skoropis) written in Cyrillic script, 14th–17th centuries. Training material: 33 pages (3,532 lines, 9,155 words) of manual ground-truth transcriptions from the record books of the Lithuanian Metrica, the archive of the Chancellery of the Grand Duchy of Lithuania (including Book of Records No. 47). The documents are administrative and legal records: copies of grand-ducal charters, privileges, land grants, court decisions and other chancery records. Language: Ruthenian (Old Belarusian), the chancery language of the Grand Duchy of Lithuania, with a strong admixture of Polish administrative vocabulary, Latinisms, and Lithuanian personal and place names rendered in Cyrillic. Script: the model works best on chancery cursive (skoropis), with connected letters, extended ascenders and descenders, superscript letters and abbreviations under titlo. It is not intended for Latin-script (Polish or Latin) entries, which also appear in the Metrica. Transcription conventions: diplomatic; superscript letters are brought down into the line; abbreviations are kept as written; historical Cyrillic letters (ѣ, і, ѳ, ъ, etc.) are preserved.

Try this model

Drag an image here

Select a file...

PNG or JPG up to 10 Mb

Wolpi
AI Assistant

By uploading an image, you accept our terms and privacy policy.

 Cyrillic handwriting_14th-17th_C
Use this modelOpen in Transkribus
Words9,155
Lines3,532
Training Pages33
Model ID668121
Languages
BelarusianUkrainian
Centuries
14th c.15th c.16th c.17th c.