OCR / Cleanup Scans
Make scanned PDFs searchable with real OCR, free and private.
Drag & drop files here, or click to select
Processing Options
How to OCR a Scanned PDF Online
Upload your scanned PDF document.
Choose the OCR language and any image cleanup options.
Click "Process Scanned PDF" — your browser downloads the OCR engine once, then recognizes each page.
Download your searchable PDF, plus a .txt file of everything that was recognized.
Why Use SonuPDF for OCR?
Real Text Recognition
Uses Tesseract, an open-source OCR engine, to genuinely recognize characters — not just re-extract text that was already selectable.
Image Cleanup Filters
Boost contrast and strip background noise before recognition to make faint or low-quality scans easier to read correctly.
Truly Searchable Output
Embeds an invisible, correctly-positioned text layer over your scan — the page looks identical, but you can now select, copy, and search it.
100% In-Browser
The OCR engine runs on your device via WebAssembly. Your PDF is never uploaded — only a generic language file is downloaded once.
Frequently Asked Questions
Real OCR. SonuPDF runs Tesseract, an open-source text recognition engine, entirely in your browser to genuinely recognize characters in your scan — it is not simply re-extracting text that was already there.
No. Recognition runs locally in your browser via WebAssembly. The only thing downloaded is a small, generic OCR language file — never your document.
English, Spanish, French, and German individually, or a combined multi-language mode that recognizes any of the four in one pass.
No — the recognized text layer is invisible. Your PDF looks the same as the (optionally cleaned-up) scan; the only difference is that its text is now selectable and searchable.
Yes, uncheck "Extract text (OCR)" to keep only the image cleanup — contrast and noise removal — without running text recognition.