OCR — scans to text
Turn scanned PDFs and photos into copyable text, or into a searchable PDF — in 14 languages, with no page limit, and nothing uploaded.
How to OCR a document
- Add a scanned PDF or photos. Clear, straight, well-lit scans recognize best.
- Pick the document's language and the output: plain text, or a searchable PDF that looks identical but is copyable.
- Run OCR. Progress is shown per page; the result downloads when done.
Frequently asked questions
How accurate is it?
On clean 300-DPI scans, very accurate. Accuracy drops with blur, skew, handwriting and low contrast — the Clean Up Scan tool can improve difficult scans before OCR.
Which languages are supported?
English, Arabic, French, German, Spanish, Italian, Portuguese, Dutch, Turkish, Russian, Ukrainian, Polish, Hindi, Chinese (simplified) and Japanese. Choosing the right language matters a lot for accuracy.
Why is it slower than server OCR?
Because it's your device doing the work instead of a data center — that's the price of your documents never leaving your machine. A typical page takes a few seconds.
Is there a page limit?
No — process 300 pages if you like. Most online services cap OCR at 100 pages or put it behind a paywall.
A note on non-Latin scripts
Text extraction (.txt) works for all 15 languages. The invisible text layer in the searchable PDF output works best for Latin-script languages; for Arabic, Hindi, Chinese and Japanese we recommend the plain-text output.