Loading Crafex...
Loading Crafex...
Scanned PDFs are images, not text. Crafex runs OCR right in your browser — WASM-powered tesseract.js, zero-server — then builds an editable DOCX. Your scan never leaves the device: 100% private.
Scanned PDF → OCR → editable DOCX, entirely in-browser.
100% Safe & Private – files never leave your browser
PDF → DOCX converted locally via pdf-lib — no upload, instant.
or click to browse from your device
PDF → DOCX • No signup
PDF → DOCX • Keep formatting — 100% private, processed locally.
Scan at 300dpi grayscale first — OCR accuracy climbs sharply on clean pages.
Run page 1 only as a test before committing a long document.
Fix obvious character confusions (O/0, l/1) in Word after conversion.
Source → Target
PDF → DOCX
Lossless where possible
Engine
Browser-Based
pdf-lib / pdfjs-dist / jszip
Privacy
100% Private
Never leaves your device
Everything for scanned pdf to word — ocr, free.
Yes — OCR extracts the text locally, then we build a DOCX you can edit in Word.
Never — OCR runs 100% client-side via WASM in your browser, zero-server processing, nothing is stored.
High on clean 300dpi prints; faint or skewed scans improve after a quick re-scan.
Free, unlimited, no sign-up, no watermark.
Explore related conversions — all private, same instant widget.