Alexander James O'Neill (Musashino University) · PyLaia · Published September 1, 2026
Pracalit to IAST
Text Recognition
Description
This model transcribes manuscripts in the Pracalit script into IAST. It is primarily tailored for Sanskrit manuscripts originating from Nepal, but it can also be utilised with Nepal Bhasa manuscripts. Training and validation data were sourced from open-source images of manuscripts dating approximately from the 14th to the 20th centuries. Transcriptions were performed and verified through manual input by Alexander O'Neill (Musashino University). The majority of manuscripts within the training dataset are in scriptio continua; consequently, the output will predominantly be in scriptio continua.
Try this model
Drag an image here
Select a file...PNG or JPG up to 10 Mb
Wolpi
AI Assistant
By uploading an image, you accept our terms and privacy policy.
Use this modelOpen in Transkribus
Low error rate9.91% CER
Character Error Rate (CER) measures the percentage of characters incorrectly recognised. Lower is better. This model scored 9.91% 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.
Words13,402
Lines1,448
Training Pages202
Model ID628689