moritz.draschner · PyLaia · Published January 26, 2026

EVMsaurus minimus

Text Recognition

Description

Under the supervision of Moritz Draschner, this model was trained on a corpus of Old English hagiography written in English Vernacular Minuscule. This hand, which survives in the 11th-century British Library manuscript Julius E.vii, was transcribed by a group of undergraduate students at the University of Düsseldorf over the span of four months. Version 1.0 of the model can reliably read square-influenced English Vernacular Minuscule hands resembling the one in Julius E.vii. It was trained to recognise unusual letterforms such as thorn, eth and wynn, as well as Tironian characters and scribal abbreviations prevalent in Old English prose and poetry. The transcription team consisted of Lina Doleys, Quinn Grabowski, Audrey Heimann, Sean McCormick-Silex, Andrej Meysing, Teona Sandu, Hanna Wierutsch, Marie Witzke, Mats Zimmermann and Jessica Zipfel. The manuscript used in the training of this model was digitised by the British Library, and the images are available in the Public Domain.

Try this model

Use this modelOpen in Transkribus
Very low error rate4.96% CER

Character Error Rate (CER) measures the percentage of characters incorrectly recognised. Lower is better. This model scored 4.96% 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.

Words24,541
Lines2,890
Training Pages94
Model ID498477
Languages
English Old (ca.450-1100)
Centuries
10th c.11th c.12th c.