Jochre 3 and the Yiddish OCR corpus
This work addresses the need for accurate OCR tools and searchable digital archives for Yiddish language materials, benefiting researchers and cultural preservation efforts, though it is incremental as it builds on existing OCR methods.
The authors tackled the problem of optical character recognition (OCR) for Yiddish texts by developing Jochre 3, an open-source tool suite, and creating a publicly available Yiddish OCR corpus. The result was a CER of 1.5% on their test corpus, outperforming existing public models for Yiddish.
We describe the construction of a publicly available Yiddish OCR Corpus, and describe and evaluate the open source OCR tool suite Jochre 3, including an Alto editor for corpus annotation, OCR software for Alto OCR layer generation, and a customizable OCR search engine. The current version of the Yiddish OCR corpus contains 658 pages, 186K tokens and 840K glyphs. The Jochre 3 OCR tool uses various fine-tuned YOLOv8 models for top-down page layout analysis, and a custom CNN network for glyph recognition. It attains a CER of 1.5% on our test corpus, far out-performing all other existing public models for Yiddish. We analyzed the full 660M word Yiddish Book Center with Jochre 3 OCR, and the new OCR is searchable through the Yiddish Book Center OCR search engine.