A groundbreaking project based on the National Library of Israel’s digital database of Hebrew manuscripts has enabled automatic transcription of the entire Cairo Geniza, making it searchable and accessible worldwide.
On Monday, November 24, 10:00-11:00, MiDRASH team leaders Professor Daniel Stökl Ben Ezra, Professor Nachum Dershowitz, Dr. Avi Shmidman, will present their work their work in automatic transcription of the Cairo Geniza and will also publish the models developed for transcribing the Geniza.
The presentation is part of a “Transcribathon” for the transcription of Hebrew manuscripts. On November 24-27, volunteers will “hack” a database of digitized Hebrew language manuscripts that have undergone automatic transcription. Their task will be to improve these transcriptions by reviewing and correcting the results of the automatic process, helping to train and enhance machine-learning models that will ultimately enable full text search of handwritten documents.
The “Transcribathon” will be held both at the National Library of Israel in Jerusalem and online, from November 24–27, between 10:00 and 18:00 (Israel time). The focus will be on semi-cursive and cursive medieval and modern Hebrew scripts. Activities will include reviewing and correcting automatic transcriptions, with full guidance provided for all participants. To participate:
https://www.nli.org.il/en/visit/seminars/manuscripts-hackathon