Scan to a searchable PDF, with text you can copy
A scanned PDF is a picture: you cannot search it, select a sentence or copy an address out of it. This page opens the scanner with Searchable text (OCR) already on, so the words on each page are recognised on your device and laid invisibly over the picture. The PDF looks exactly like the scan, but a search finds the words, the text can be selected and copied, and desktop search indexes it. The recognition runs in your browser; the pages are not uploaded.
Scan a document
Use the camera or choose photos of documents (JPG, PNG, WebP, HEIC). Everything stays on this device.
Hold the phone straight above the page
Choose the language of the document before saving: the engine knows the words and the letter shapes of that language, and the accuracy drops sharply with the wrong one. A second language can be mixed in for an invoice with English terms. Fifteen languages are available (English, Romanian, Spanish, Portuguese, French, German, Italian, Dutch, Polish, Hungarian, Turkish, Czech, Russian, Ukrainian and Greek); the language data is fetched from this site once and kept on the device.
The text layer is verifiable: open the PDF in any reader, search for a word from the page and it is highlighted at the right place. Printed text only: handwriting is not recognised.
Tips
- Straight, sharp, evenly lit pages give the best text: fix the corners and the filter first, then save with OCR.
- Black & white is usually the most readable filter for the engine on plain text; Natural for pages with colour or stamps.
- Recognition takes a few seconds per page on a phone; the progress shows which page it is on, and you can cancel.
- To get the text out as a file rather than a PDF, run the saved PDF through Image to text (OCR).
- Redact before saving if a page has data the recipient should not search for: the words under a redaction are removed from the text layer too.
Why it is private
Everything you type or drop here is handled by your browser, on your device. Nothing is uploaded. To verify, open your browser's developer tools, go to the Network tab, and use the tool: you will see no request carrying your data.
Frequently asked questions
- Is the text sent to a server?
- No. The engine (Tesseract) and the language data run in a Web Worker in your browser; the pages never leave the device.
- How accurate is it?
- On a clean, straight page in the right language, printed text comes out almost entirely right; small print, dim photos and wrong languages give wrong letters. Check by searching for a word.
- Can I edit the recognised text?
- Not in the PDF: the layer is invisible text under the picture. For editable text, use Image to text (OCR), which gives a .txt you can edit.