Emanuel Elyasaf · PyLaia · Published February 7, 2024

Pinkas Brody Model

Text Recognition

Description

Pinkas Brody – the community notebook of Brody (formerly in east Gallica, now in western Ukraine) contains various agreements between members of the Brody community between the years 1807–1817, including marriage agreements, inheritance, partnerships and dissolution of partnerships, sale and rental of houses, seats in the synagogue, and more. The handwriting is that of the community scribe and generally is clear and legible. The pinkas was deciphered by Emanuel Elyasaf during 2023. At the request of Ms. Marlis Glaser Humphrey, Dr. Sallyann Amdur Sack, and under the guidance of Mr. Jan M. Gronski all from L'Dor V'Dor and IAJGS, Emanuel uploaded the deciphered pinkas to Transkribus and built a model for Hebrew and Yiddish. The pinkas comprises 158 pages with approximately 180,000 words. The text recognition model was built with a validation set of 10 percent of the trained data. Its CER and WER are 4.4 percent and 15.3 percent respectively.

Try this model

Pinkas Brody Model
Use this modelOpen in Transkribus
Very low error rate4.4% CER

Character Error Rate (CER) measures the percentage of characters incorrectly recognised. Lower is better. This model scored 4.4% 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.

Words159,013
Lines12,670
Training Pages143
Model ID59324
Languages
HebrewYiddish