Bentham Project (University College London), DEEDS-project (University of Toronto) · PyLaia · Published December 13, 2022

UCL–University of Toronto #7

Text Recognition

Description

Seventh iteration of the collaborative UCL–University of Toronto model for processing medieval Latin manuscripts, particularly those containing a large quantity of abbreviated words. E-mail: c.riley@ucl.ac.uk.

Try this model

Use this modelOpen in Transkribus
Very low error rate1.7% CER

Character Error Rate (CER) measures the percentage of characters incorrectly recognised. Lower is better. This model scored 1.7% 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.

Words140,158
Lines10,303
Training Pages330
Model ID48734
Centuries
13th c.14th c.15th c.