konstantin.gromyko · PyLaia · Published July 15, 2024

Russian and Ukrainian handwriting XXI century

Text Recognition

Description

History of railroads in Russia (2024) Digitized books in Russian and Ukrainian languages. Transkribed with help of democratic society community in Helsinki, Finland. Purpose of the model: to enable legal - FOR PRIVATE USE only - sharing of books as fair use exception from copyright laws in certain EU MSs (with Public Lending Right (with CMO in place)), in independent secure digital lending mode : one copy to one user, (i)SDL. E.g per Finnish Copyright Act, Section 12, private copying exception from copyright limitations.

Try this model

Use this modelOpen in Transkribus
Low error rate8.65% CER

Character Error Rate (CER) measures the percentage of characters incorrectly recognised. Lower is better. This model scored 8.65% on its validation set. As a rule of thumb, a CER below 10% is considered good for most handwritten material. This is a smaller, specialised model. It may achieve a very low CER on material similar to its training data, but could be less robust on unfamiliar handwriting or layouts.

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.

Words11,363
Lines2,452
Training Pages46
Model ID132853
Languages
RussianUkrainian
Centuries
21st c.