How to oCR: make a scanned PDF searchable
Add your scanned PDF
Choose a PDF made by a scanner or phone, or drag it onto the page.
Recognise the text
Each page is read by the text recognition engine in your browser. This takes a few seconds per page.
Download
Get a searchable PDF that looks the same as the original, plus the text on its own if you want it.
About this tool
A scanned document is a picture of a page: you can't search it, select a sentence or copy a reference number. Optical character recognition (OCR) reads the picture and finds the words. They're added to each page invisibly, in the right place, so the PDF looks unchanged but behaves like a normal document.
Recognition runs entirely in your browser using Tesseract, a well-established open-source OCR engine. That keeps private letters, bank statements and ID documents on your device. It works best on clear, straight scans of printed English text. Handwriting and very blurry pages are hard for any OCR engine to read, so always check important details such as numbers and names.
Questions
Is my document uploaded?
No. The recognition engine is downloaded to your browser from our site, and your pages are read on your own device.
Which languages are supported?
English for now. Other languages will be added; each one needs its own recognition model.
Will the PDF look different?
No. The original pages are kept exactly as they are. The recognised words are added as invisible text over the matching places.
How long does it take?
Usually a few seconds per page on a computer, and longer on a phone. The first time, the recognition engine (about 9 MB) downloads once.
Can it read handwriting?
Only very neat handwriting, and not reliably. It's designed for printed text.