OCR
Recognize text from scanned or image-based PDFs, entirely in your browser.
Drop your PDF here
or click to browse
Up to 10 MB · 20 pages
How to ocr a scanned pdf
OCR (optical character recognition) reads text out of scanned pages and photographs, turning pictures of words into actual text you can search and copy. The recognition engine runs in your browser.
- 1
Choose your scan
Drop the scanned PDF or image you want recognised.
- 2
Choose the language
Select the language of the document. Accuracy depends heavily on this being correct.
- 3
Run and download
Click Run OCR. Recognition takes a little time as it works page by page, then the text is available to copy or download.
Your files never leave your device
Unlike most online PDF tools, this one uploads nothing. Everything runs inside your browser, so your document is never transmitted, never stored on a server, and never seen by anyone but you. Once the page has loaded you can disconnect from the internet entirely and the tool still works.
Frequently asked questions
- Are my scans uploaded anywhere?
- No. The OCR engine itself is downloaded to your browser and runs there. Your scans are never transmitted, which is unusual for OCR tools and matters for sensitive documents.
- Why is OCR slower than other tools?
- Character recognition is genuinely heavy computation, and it is happening on your device rather than on a server farm. Larger documents take proportionally longer.
- How accurate is it?
- Clean, straight, high-resolution scans of printed text read very well. Handwriting, low-resolution photographs, and skewed pages are considerably less reliable.
- Which languages are supported?
- Many, including Latin, Cyrillic, Devanagari, Arabic, and CJK scripts. Choose the language before running, as the wrong selection produces poor results.
Related tools
DinoPDF — every tool runs in your browser. Nothing is uploaded.