Tools › PDF tools
OCR: make a scanned PDF searchable
Recognise the text in scanned PDFs and photos of documents. Get a searchable PDF that looks exactly like the original, plus plain text to copy. Runs on your device.
How it works
- Each page is rendered at 300 DPI and read by Tesseract, the open-source OCR engine, running in your browser.
- The recognised words are laid invisibly over the original page, so the PDF looks the same but you can search, select and copy its text.
- Pages that already contain text are skipped, so mixed documents are not processed twice.
- The engine and language models (about 2–3 MB per language) download from this site on first use and then work offline.
Questions
Are my documents uploaded?
No. Recognition happens on this device; only the OCR engine itself is downloaded. That makes it safe for passports, bank statements and medical records.
Which languages are supported?
English, Spanish, Portuguese, French, German, Italian, Polish, Turkish, Russian, Ukrainian, Uzbek (Latin and Cyrillic), Kazakh, Indonesian, Vietnamese, Thai, Hindi, Bengali, Urdu, Arabic, Persian, Chinese (Simplified and Traditional), Japanese and Korean. Choose up to three at once.
How accurate is it?
Clean printed text at 300 DPI is usually read almost perfectly. Handwriting, low-resolution photos and decorative fonts are much harder; if the result looks wrong, check the language and that the page is upright.
Does it change how the PDF looks?
No. The original page stays untouched; the text layer is invisible.