Thursday, September 19, 2019

Qoruyo: Models for Automatic Transcription of Manuscripts

Qoruyo: Models for Automatic Transcription of Manuscripts 
The Beth Mardutho Qoruyo project seeks to develop tools and resources for successful optical character recognition (OCR) and handwritten-text recognition (HTR) of printed and handwritten Syriac texts.
The project is now pleased to announce the Beth Mardutho Transkribus HTR models, which can automatically transcribe handwritten Syriac documents written in Estrangelo, East Syriac, and Serto, with up to 98% accuracy.
Acknowledgements
These models were trained as part of the Beth Mardutho Summer Fellowship Program 2019.The Estrangeloand Serto models were trained by Kyle Brunner (NYU), and the East Syriac model was trained by Abigail Pearson (University of Exeter).
What is Transkribus?
Transkribus is a free software program which can automatically transcribe handwritten text from digital images using HTR technology.  Completed transcriptions can then be edited, searched, tagged, and exported using tools available in the program.   
Using Transkribus, we have trained the following three HTR models to recognise handwritten Syriac in all three scripts with impressive accuracy:
Qoruyo Estrangelo Beta 1.0 – up to 98% accuracy
Qoruyo East Syriac Beta 1.0 – up to 95% accuracy
Qoruyo Serto Beta 1.0 – up to 96% accuracy
These models are accessible to anyone with a Transkribus account and, as they were trained using data from several manuscripts, they can be used successfully on a variety of handwriting styles.  

No comments:

Post a Comment