Make a PDF searchable
Scanned pages are pictures of words, so you cannot search them or copy from them. This reads the words, using text recognition that runs on your own device, and adds them invisibly behind each page.
How to use it
- Choose a scanned PDF, or a photo of a document.
- Press Make searchable. Each page takes a few seconds.
- Download the searchable PDF. It looks the same, but now you can search and copy the text.
How it works
Each page is read by Tesseract, a well-established text recognition engine, running inside your browser. The words it finds are placed in an invisible layer exactly behind the picture of each page, so the document looks unchanged but you can search it, select text and copy it. Pages that already have text are left alone unless you ask otherwise.
The first time you use it, your browser downloads the recognition engine and its English language data, about 9 MB in all, from this website. Your document itself never leaves your device.
How accurate is it?
Clean, straight, typed pages are read very accurately. Crooked photographs, faint print, handwriting and unusual fonts are read less well. The page reports how confident the recognition was, so you know when to check the result more carefully.
Questions
Which languages does it read?
English, for now.
Can I then convert it to Word?
Yes. Use Keep going with this file to open the searchable copy in PDF to Word.
Are my files uploaded?
No. The work happens inside your web browser, on your own device. Before your file is opened, this page locks itself: from then on your browser refuses to let it fetch, upload or send anything at all. You can check: once the page has loaded, switch off your internet connection and it will still work. How the lock works.