Skip to content

OCR PDF

Recognise text in scans to make PDFs searchable or extract text.

Good to know

Hindi, Chinese and Japanese are recognised into the .txt export only (no font for the searchable layer yet). Arabic selection order may be imperfect. Accuracy depends on scan quality.

About this tool

OCR PDF reads the text in scanned pages and photographed documents, where the words are only part of a picture. You can turn the result into a searchable PDF, save the text as a plain .txt file, or do both.

How to use it

  1. Choose one or more scanned PDFs.
  2. Under “What do you want?”, pick Make searchable PDF, Extract text or Both.
  3. Tick every language that appears in the document. English is ticked to start with.
  4. Leave “Skip pages that already have text” on to keep pages with real text as they are; it is also faster.
  5. Select the button below the options and wait while each page is read on your device. How long it takes depends on the number of pages, the quality of the scan, the languages you selected and the speed of your device.

Good to know

  • Each page is rendered at 300 dpi before it is read. Very large pages are read at a lower resolution, and the result lists them.
  • Accuracy depends on the scan. Blurred, skewed or faint pages produce more mistakes, so check names, dates and figures before relying on them.
  • English, German, French, Spanish, Italian, Portuguese, Dutch, Russian and Arabic can go into a searchable PDF. Hindi, Chinese (Simplified) and Japanese are recognised into the .txt file only.
  • The recognition engine and language data are loaded from this site the first time you use them; your document itself stays on your device.
  • Your files are processed in this browser tab and are never uploaded to a server. How your privacy is protected

Common questions

Does OCR change how my pages look?

No. The searchable PDF keeps each page image as it was and adds an invisible text layer behind it, which is what you search and copy.

Can I make a password-protected PDF searchable?

Not while it is protected. Remove the password with Unlock PDF, then run OCR on the unlocked copy.