About OCR PDF
OCR PDF reads scans with a recognition engine and turns them into text you can search, copy and cite. Pick the document's language for the best accuracy and download the result as plain text.
The engine core and language data load once from a CDN and are then cached by your browser, so the second run is much faster than the first.
How to use OCR PDF
- Open your scanned PDF.
- Choose the document language.
- Click Run OCR, then download the text.
Six languages
English, Indonesian, Spanish, Portuguese, French and German.
Page markers
Output is divided by page, so citations stay possible.
Local processing
Recognition runs in your browser; the scan is not uploaded.
Honest expectations
OCR is strong on clean prints at 200 DPI or better and weaker on handwriting, skew and low contrast. Names, numbers and tables deserve a manual check. The result is a text file: to rebuild a searchable PDF you would pair it with the original scan, which full OCR suites do. For text-based PDFs that just need exporting, PDF to Text is instant.
Tips for better results
- Scanned at a tiny size? The page is rendered at 200 DPI regardless, which usually helps.
- Multi-language document? Run it once per language and merge.
Questions and answers
Does it need internet?
Once, to fetch the engine and language data. After that the browser cache takes over.
Why is the first run slow?
It downloads the recognition engine and language data; later runs reuse them.