OCR — make scans searchable
Turn a scanned PDF or image into a searchable PDF and pull the text out — accurate 300-DPI recognition, handled honestly: pages are prepared in your browser, read on our server, then deleted immediately.
OCR is a Pro tool
Reading text off scanned pages is the one job here that cannot happen in your browser — it runs on our server and takes real time per page. That is why it sits behind Pro while everything that can run on your device stays free and unlimited.
Only need the text, not a searchable PDF? If your PDF already has a text layer, PDF to Text pulls it out free, in your browser.
0 bytes of your file have left this device. Counted live in this tab, not claimed — open your browser’s Network tab and watch.
How it works
- 1
Drop in a scanned PDF or a photo/image of a document.
- 2
Your pages are sharpened to 300 DPI in your browser, then read by our OCR engine.
- 3
Download a searchable PDF (with a selectable text layer) plus the extracted text — your file is deleted immediately.
Frequently asked questions
What does OCR do?
OCR (optical character recognition) reads the text inside a scan or photo, which is otherwise just a picture. You get back a PDF whose text you can select, copy, and search — plus the plain text on its own.
Why does this tool use a server?
Accurate text recognition uses a full OCR engine (Tesseract) that runs best on a server. To keep it private, your pages are prepared in your browser first, sent over an encrypted connection, read, and then deleted immediately — nothing is stored or logged. See the “Where your data goes” table on our Security page.
Does the layout stay the same?
Yes — the searchable PDF keeps your original page image exactly as-is and adds an invisible, selectable text layer on top. So it looks identical, but the text is now searchable and copyable.
Which languages are supported?
Over 100 — from English, Spanish, French, German and Chinese to Arabic, Hindi, Japanese, Russian, Thai and many more. Pick your document’s language before running (it improves accuracy). Clear, straight scans work best; very faint, skewed, or handwritten pages are harder for any OCR.
How big a document can it handle?
Large ones. Pages are prepared in your browser and streamed to the server in small batches, so a 100-plus-page, tens-of-megabytes PDF works fine without overloading anything — up to 500 pages per job. You can also OCR just a page range, and choose Best (300 DPI, most accurate) or Fast (200 DPI) quality.
What usually comes next
The step most people take after this one — and why.
Last improved
Changes to this tool specifically, not the site.
- New tool
OCR is open — read the text off a scan
Point it at a scanned PDF or a photo of a page and you get the document back with real, selectable, searchable text underneath — the pages look exactly as they did, and the file barely grows, because we lay an invisible text layer over your originals rather than re-rendering them. Over fifty languages, and you can pick a page range. This one is Pro: it is the single job here that cannot run inside your browser, so it runs on our server and takes real time per page, which is why everything that can run on your device stays free and unlimited. If your PDF already has a text layer and you only want the words out, PDF to Text does that free.