Image to text

Turn a picture of text into real, usable text

A photo of a whiteboard, a screenshot of an error message, a picture of a printed page — the words are right there on screen, but you can't select or copy a single one of them, because to a computer it's just an image. Optical character recognition, OCR, looks at the shapes of the letters and turns them back into real text, the same technology behind a phone's document scanner app.

How it works

Drop an image and a text recognition engine, running entirely on your device, reads the letters and words it can find in it. The result appears as plain text you can copy, edit or download, ready to paste anywhere.

The first time takes a little longer

The recognition engine and its language data, a few megabytes, download the first time you use this tool or the PDF OCR tool, and your browser keeps them after that. That makes the very first image slower than the ones after it.

Getting a good result

Accuracy depends heavily on the source: a sharp, well-lit, straight-on photo with good contrast reads far better than a blurry or tilted one. Printed text is read much more reliably than handwriting, and some mistakes should be expected on any image, so check the result rather than trusting it blindly for anything important.

Privacy

Your image is read entirely in your browser. It's never uploaded anywhere, which matters for screenshots that often contain private conversations or personal information.