Recognize English text in scanned pages
Run OCR to add an invisible searchable English text layer to rendered PDF pages.
How it works
- Open a scanned PDF with legible English text.
- Run recognition. The first use downloads a recognition model, so an internet connection is needed for that download.
- Export the searchable PDF and check names, dates and numbers by searching or copying text.
The first run downloads a recognition model. Accuracy varies; review names, numbers and important text. Other languages are not currently supported.
When to use this tool
Make scanned notes or documents searchable. Recognition adds an invisible text layer to rendered pages.
What to check before sharing
Recognition accuracy depends on scan quality. English is currently supported. OCR does not recreate a Word document or make image text directly editable.
Supported files
Downloads an English recognition model on first use. Accuracy varies; review important text.
Inputs: PDF. Performance depends on your device. Download important work before closing the browser; local storage is not a backup.
Troubleshooting
If the file will not open, check it in another PDF reader. If it is password-protected, use Remove a PDF password with a password you know. For scans that need selectable text, run Make a PDF searchable first.
Common questions
Does OCR upload my PDF?
Recognition runs in the browser; model assets are downloaded separately.
Are other languages supported?
This implementation currently supports English. Review recognized text before relying on it.
Which files can I use?
Supported inputs: PDF. Downloads an English recognition model on first use. Accuracy varies; review important text.
Is this free?
Core tools require no account or subscription and add no watermark.
Do my documents leave my device?
Document processing runs in your browser. PDF Studio does not upload document files. Explicitly saved drafts and My files remain in this browser until deleted. Read aloud may use network-backed browser voices.