OCR a scanned PDF — on your device

Got a scan or a photo of a document and need the text out of it? Run free OCR that works right in your browser. The recognition engine downloads once, then reads your pages locally — the file itself is never uploaded.

How to OCR a scanned PDF

1. Open the OCR tool and add your scanned PDF or image (JPG/PNG).
2. Choose the language of the document for the best accuracy.
3. Run OCR. The first run downloads the language model (~2–12 MB); after that it's cached.
4. Download the text that was recognised.

Text-based vs scanned PDFs

A text-based PDF already contains real, selectable text — for those, the faster Extract text tool is all you need. A scanned PDF is really a picture of a page, so selecting or copying gives you nothing. OCR (optical character recognition) looks at the picture and works out the letters. If Extract text comes back empty, your PDF is scanned — use OCR.

Getting the best results

• Straight, well-lit scans read far better than skewed phone photos.
• Pick the correct language — it changes which characters the engine expects.
• Higher-resolution scans (300 DPI) give more accurate text than low-res ones.

Frequently asked questions

Is my scan uploaded to a server?

No. The engine downloads to your browser; your file is read on your device and never uploaded. See our privacy policy.

Which languages are supported?

English, Greek, Russian, French, German, Spanish and Italian.

My extracted text was empty — why?

Your PDF is scanned (a picture of text). Run OCR to turn the picture into text.