ephremishac · PyLaia · Published January 12, 2026
Syriac New Testament of Jerusalem (Estrangelo - Melkite)
Text Recognition
Description
Model trained on the manuscript: MS 36, Saint Mark's Monastery - Jerusalem. The second public Syriac model on Transkribus was prepared by the Syriac Group in IMAFO during the HTR Winter School 2025 - Vienna. Supervisors: Ephrem Aboud Ishac (Vienna, IMAFO) & Christine Roughan (Princeton, MARBAS). Participants: Julius Balski, Nima Jamali, Jeanette Kilicci, Chia-Wei Lin, Andrei Macar, Branko Malešević, Stefan Marić, Daniele Reano, Rita Sawaya, Jonathan Stutz. With gratitude to the Syrian Orthodox Monastery of St Mark - Jerusalem. The manuscript is made available courtesy of vHMML (SMMJ 00036):: https://www.vhmml.org/readingRoom/view/126351
Try this model
Use this modelOpen in Transkribus
Very low error rate3.53% CER
Character Error Rate (CER) measures the percentage of characters incorrectly recognised. Lower is better. This model scored 3.53% 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.
Words53,225
Lines15,963
Training Pages119
Model ID468925