# Searchable OCR — add a selectable text layer, no upload

> Run on-device OCR over your scanned pages and export a PDF with an invisible, selectable text layer. The Tesseract engine runs in your tab — nothing uploaded.

Run OCR over your scanned pages and export a PDF with an invisible, selectable text layer over the image — so the document is searchable and copyable. The Tesseract engine runs in your browser tab, so pages are never uploaded.

A scan is just a picture until it has a text layer — you can't search it, copy from it, or have a screen reader read it. OCR fixes that by recognizing the characters and laying them invisibly over the image, so the page looks the same but behaves like a real document.

Accuracy tracks input quality, which is why the enhance steps matter: straight, high-contrast pages recognize far better than dim, tilted ones. Choosing the right language further sharpens the result.

Clean pages first with Deskew and Scan filters, and if you have an existing PDF rather than a scan session, the Document Desk's OCR handles that too.

## How to

1. **Prepare pages.** Scan or import pages and clean them up for accuracy.
2. **Run OCR.** Tesseract recognizes the text on each page in your browser.
3. **Get a text layer.** The recognized text is placed invisibly over the image.
4. **Export.** Download a searchable PDF you can select and search.

## FAQ

### What does OCR add to my scan?

It adds an invisible text layer aligned to the page image, so the PDF looks identical but you can search, select, and copy the text — and screen readers can read it.

### How do I get the best OCR accuracy?

Clean input helps most: de-warp and deskew the pages and use the black & white filter so text is crisp before running recognition.

### Which languages are supported?

Tesseract supports many languages; pick the one matching your document so characters and diacritics are recognized correctly.

### Is my document uploaded for OCR?

No. Recognition runs via WebAssembly in your browser, which is what lets you OCR confidential paperwork without it leaving the device.


## Related tools

- [Deskew](https://www.safepaper.app/scan/deskew)
- [Scan filters](https://www.safepaper.app/scan/filters)
- [OCR](https://www.safepaper.app/docs/ocr)
- [Build PDF](https://www.safepaper.app/scan/to-pdf)

---

Canonical HTML: https://www.safepaper.app/scan/ocr
Markdown: https://www.safepaper.app/scan/ocr.md

There is no upload endpoint — your files are processed in this browser tab. Open your network tab and check.