OCR PDF

How it works

1. Choose a scanned PDF.
2. Choose the document language and an output: a searchable PDF or a text file.
3. Run OCR and download the result. The first run takes longer while the language model loads.

OCR, or optical character recognition, reads the text in a scanned page. This tool runs Tesseract in your browser, so the scan is not uploaded. English, Spanish, French and German are supported.

Searchable PDF or text file

A searchable PDF keeps the scan exactly as it looks and adds an invisible text layer underneath, so you can search, highlight and copy text. A text file contains only the recognized text, which is useful for editing or pasting into another program.

Pages that already have text

Pages that already contain selectable text are read from that text. Only pages without it go through OCR, so a mixed document takes less time than a fully scanned one.

Getting better results

• Scan at around 300 DPI. Low resolution and blurry photos reduce accuracy.
• Straighten skewed pages. Recognition works best on upright text.
• Pick the language that matches the document. Text that mixes languages is recognized less accurately.
• Handwriting is not reliably recognized.

Limits

This tool processes your file on your device and never uploads it. The app does not set a size limit for it, so very large files can run out of memory in the browser, especially on phones. If that happens, split the file first with Split PDF.