Word Lens
Tap source words to see the word, unvowelized form, transliteration, sentence context, likely meaning, and notes.
Current beta: v1.0.3 build 82
Scan or import Torah and Chassidus pages, translate them with your chosen provider, review source-linked footnotes, listen in Reader Mode, and export full projects for learning, teaching, and research.
Built around the study workflow
Create a project for a sefer, maamar, manuscript, PDF, or scan set. Add pages, preview corpus PDFs, translate one page or the whole project, review footnotes, tap words for context, and export the result.
Capture a page, import a PDF, or browse the Mafteiach and Chabad Library corpus before selecting pages.
Select translation provider, model, template, output language, OCR mode, and batch concurrency before a run.
Inspect footnotes and sources, use Word Lens on source words, listen in Reader Mode, then export pages or full projects.
Latest beta highlights
Build 82 adds tap-to-understand Word Lens study help, keeps the current model/run controls visible, and continues hardening the Reader PDF/export workflow.
Tap source words to see the word, unvowelized form, transliteration, sentence context, likely meaning, and notes.
Browse Mafteiach and Chabad Library entries, preview PDFs, select pages, batch translate, and export full projects.
Reader PDF exports are checked with local fixture regression before release so layout issues are caught earlier.
Who it is for
Turn a page you are studying into readable English, while preserving the source context and footnotes.
Build project-based translations, review provider output, generate summaries, export reader PDFs, and prepare material for shiurim or handouts.
Offer a guided station near the stacks where patrons can scan, translate, review, and contribute text-only translations for review.
Community library
The Global Library lets users submit completed text translations for private review. Approved contributions become browsable by work, topic, source, and weighted Torah-learning tags.
Original scans, API keys, local file paths, precise location, and downloaded model/library files are not published.
Submissions wait in private review before they appear publicly in the Global Library.