OCR a PDF online — free, no upload

A scanned PDF is just pictures of pages: you can't select or search the text. OCR (Optical Character Recognition) fixes that by reading the characters from the image. This tool runs Tesseract OCR entirely in your browser — your document is never uploaded to any server.

How the OCR works

• Automatic detection — each page's native text layer is checked first; only pages that need it go through OCR, so digital PDFs stay instant.
• High-resolution recognition — pages are rendered at ~2400 px on their longest side before recognition, which significantly improves accuracy on small print.
• Confidence scores — every OCR'd page is labeled with its recognition confidence, so you know which pages to double-check.
• Force-OCR mode — if a PDF has a corrupted or garbled text layer, switch the mode to "Force OCR" to re-read every page from pixels.

Supported OCR languages

English, Français, Español, Deutsch, Italiano and Português — including a combined English + French model, useful for mixed documents.

Tips for better OCR results

1. Prefer scans at 300 DPI or higher; blurry photos lower accuracy.
2. Pick the right recognition language before dropping the file.
3. Check pages with a low confidence score against the original.

Want the whole workflow? See the guide to extracting text from a PDF, or learn about the PDF to text converter.