A scan or a photo of a document is only a picture: you cannot search it, copy a number from it or have it read aloud. OCR reads the words and lays them invisibly over the page, so the file looks exactly the same but behaves like a real document. It runs in your browser, in English, Hindi, Marathi, Gujarati, Bengali, Punjabi, Tamil, Telugu, Kannada, Malayalam and Urdu.
Files stay on your device. Free, with no sign-up.
Nothing left your device.
Old bank statements and certificates that exist only as scans, a book chapter photographed page by page, a letter someone sent as a picture, a signed form you need to quote from. Once the text is in the file you can search a hundred pages for one name, copy an account number without retyping it, and run Redact PDF's search over the whole document.
Start from the clearest scan you have: straight, evenly lit, 200 to 300 dpi. If the pages came from a phone, run them through Scan to PDF first. Tick only the languages that are really on the page, because every extra language makes the engine slower and a little less sure.
No. Everything happens inside your browser, on your own device. Your files are never sent to a server, and you can confirm that in your browser's network tab. Close the tab and they're gone.
No. The scan is left exactly as it is, and the recognised words are added as an invisible layer on top, in the same places as the words in the picture. The file barely grows. You can search it, select text, copy it and have a screen reader read it.
On a clean, straight, well-lit scan of printed text it is very good, usually better than 95% of the words. It is weaker on handwriting, photos taken at an angle, faint or crumpled paper, and very small print. The result tells you how sure it was, and names the pages that need a look. Detailed quality helps with small print, and Scan to PDF straightens a phone photo first.
English, Hindi, Marathi, Gujarati, Bengali, Punjabi, Tamil, Telugu, Kannada, Malayalam and Urdu. For a page in two languages, such as an English form filled in in Hindi, tick both. Each language is a small file of 1 to 3 MB, downloaded from this site the first time and kept on your device after that.
Reading text from a picture is a lot of work, and here it is done by your own device rather than a server. Expect a few seconds a page. A phone is slower than a computer. You can cancel at any time.
Turn it the right way up first with Rotate PDF, then read it. The engine reads text that is upright.
Yes: PDF to Word, PDF to Excel and PDF to Text have a "Read scanned pages" option that does this step for you.
Not directly. Use Unlock PDF on this site to remove the password (you need to know it), then add the unlocked copy here.
Each file can be up to 100 MB, and up to 300 MB in total. Because the work happens on your device, very large files depend on your available memory. A desktop browser handles them best.