Image to text (OCR)
Turn a photo, screenshot or scan into text you can copy and edit — read entirely on your own device, nothing uploaded.
Getting the best result from OCR
Character recognition works best on sharp, well-lit images with dark text on a light background. Photograph documents straight-on rather than at an angle, crop to just the text if you can, and prefer the original screenshot over a photo of a screen. Currently reads English and numbers; more languages are planned.
Is my image uploaded anywhere?
No — the recognition engine runs inside your browser (WebAssembly). Your image never leaves your device, which also makes this safe for documents you'd rather not share with a random website.
Why does the first use take longer?
On first use your browser downloads the reading engine and its English language data (about 7 MB) from this site. It's cached afterwards, so later conversions start much faster.
Why are some words wrong?
OCR is a best-effort guess, and blur, handwriting, decorative fonts or low contrast reduce accuracy. Always proofread the result — especially numbers.