OCR PDF
Make a scanned PDF searchable.
- Zero-transmission processing
- No watermark
- Limited daily trial
- Unlimited on Pro
Drop your PDFs here
or pick them from your device
Your files are never transmitted. Nothing is uploaded.
Good to know
- Accepts
- Files at once
- One at a time
- Pages per run
- Up to 100
- Transmission
- None — nothing is uploaded.
Chain it
Run this tool and the next one back to back — nothing to download in between.
How to do it
- 1
Add your file
Add a scanned PDF or a photographed document — the kind where you can see words but cannot select them. That is exactly what this fixes.
- 2
Set it up the way you want
Pick the language of the document. It matters more than people expect: recognition uses a language model, and the wrong one turns accented characters into noise. Pick more than one if the pages are mixed.
- 3
Download the result
Download the result. It looks the same as before, but there is now an invisible text layer behind the image, so the words can be selected, searched and copied.
Questions about OCR PDF
- Does the page look different afterwards?
- No. The original image is kept and the recognised text is placed behind it, so the document looks identical while becoming searchable. Nothing is redrawn or re-typeset.
- How accurate is it?
- Good on clean, straight, reasonably sharp scans; worse on faint photocopies, handwriting and pages photographed at an angle. If a scan is skewed, straightening it first noticeably improves the result.
- Is my scan sent to a recognition service?
- No. The recognition engine and the language data are downloaded once and run without sending anything back. Medical records, contracts and ID documents are never transmitted, which is unusual for OCR and deliberate here.