Convert documents
Make a PDF searchable with OCR
Recognise scanned text in 100+ languages and keep every page looking unchanged.
Choose a file
Choose one scanned PDF with up to 80 pages or 120 MB.
How it works
- 1
Choose a scanned PDF and select the language printed in the document.
- 2
Rebuild recognises image-only pages and preserves pages already searchable.
- 3
The Unicode text layer and every output page are independently checked before download.
Questions
Which languages are supported?
More than 100 language models in the current Tesseract registry are available, including Turkish, Arabic, Indic scripts, Chinese, Japanese, Korean and historic variants.
Will the PDF look different?
No. OCR adds invisible searchable Unicode text over the original pages instead of rebuilding their visible appearance.
Does OCR upload my pages?
No document page is uploaded. The selected recognition pack is downloaded and cached, then OCR runs in this browser.