Skip to the card

2026-11-20 issue

Observed arrival · 2026-11-20

HeOCR makes Hebrew document recognition auditable

heocr.org Visit website
Editorial interest 84/100 Selection signal · not a rating of the site

An open-source toolkit and collection of datasets and models for extracting printed and handwritten Hebrew from documents.

Landing page captured for the 2026-11-20 issue.
For
Researchers and developers working with Hebrew documents
Worth noticing
The recommended R7 model lacks a held-out historical-print test set; reference transcriptions are still being gathered from volunteers.

Field notes

The Python package handles PDFs, Word, Excel, PowerPoint, web pages, and scans, choosing a PDF text layer when it considers that layer reliable and OCR otherwise. The HASH archive is described as 198 scans from 48 sources, with material including Hebrew handwriting from the 18th century onward. The project says rights are recorded per item; it also flags that its historical-print test set is waiting for volunteer reference transcriptions.

Observed signals

Read the marks

Editorial observations of this landing page, not a rating.

○OpenPublic substance visible
✦PrettyNotable craft visible
●ProPolished or operationally mature
◎NicheUnusually specific use

One card from the complete issue

The Woven Index

386,049 arrived 1,000 judged 972 catalogued Enter the complete issue
heocr.org

Landing page observed 2026-11-20. The live site may have changed.