Ronny Reshef, Mirjam Gutschow · PyLaia · Published November 29, 2023
Vaybertaytsh.YidTakNL
Text Recognition
Description
The Vaybertaytsh typeface model Vaybertaytsh.YidTakNL was deployed to support ongoing research done at Erasmus University (2021-2026). The model and its accompanying baseline model, in conjunction with the dataset YidTakNL can facilitate reading and researching Yiddish texts printed in Vaybertaytsh (18th-19th century).
Yiddish texts were printed in the Vaybertaytsh typeface during the 16th-19th centuries throughout Europe. Vaybertaytsh is a semi-cursive Ashkenazi typeface, also called ivre-taytsh, vayberksav, taytsh, Tsene-(u)rene-ksav, kleyn-taytsh, Tkhine-ksav and mashket/mesheyt
(Spinner 2019, 152; Zafren 1982, 138; Jacobs 2005, 47; Fishman 1991, 44; Fleischer 2018, 266).
For the untrained eye, this typeface is often difficult to read.
The text recognition model and accompanying dataset can enable researchers and students working with Yiddish written in Vaybertaytsh to create a digital corpus in an efficient and precise manner.
יידיש, צאנה וראנה, אותיות צו"ר
Based on our previous model [Model ID
56218], which is based on the Dibbuk.
Try this model
Use this modelOpen in Transkribus
Very low error rate0.9% CER
Character Error Rate (CER) measures the percentage of characters incorrectly recognised. Lower is better. This model scored 0.9% 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.
Words66,497
Lines8,062
Training Pages242
Model ID57147