Image to text converter (OCR)

What OCR does

Optical character recognition (OCR) turns the pixels that look like letters in a photo or screenshot back into actual, selectable text you can copy, search and edit. It is how you lift a quote off a screenshot, capture a serial number from a photo, or digitise a scanned page without retyping it. Under the hood the engine finds the text regions, segments them into lines and characters, and matches each shape against a trained model of what letters look like — which is why it is pattern recognition, with a confidence level, rather than a perfect transcription.

Getting clean results

OCR accuracy depends almost entirely on the input. It is strongest on clear, high-contrast, roughly horizontal printed text — dark type on a light background. It struggles with handwriting, stylised or decorative fonts, low resolution, motion blur, skew, busy backgrounds, and text laid over photos. A few habits help a lot: use the sharpest image you have, crop tightly to just the text you want so the engine is not distracted by surrounding graphics, and straighten a tilted scan first. Expect to proofread the output — numbers, punctuation and the classic confusions (O/0, l/1/I, rn/m) are where errors cluster.

Runs in your browser

This tool runs Tesseract, a mature open-source OCR engine, compiled to WebAssembly so the recognition happens on your machine — your image is never uploaded. Only the engine core and the English language model are downloaded (once, then cached), so the first run is slower while they load and later images are quick; the picture itself stays in the browser. That privacy matters for screenshots of documents, IDs or private messages. To pull text out of a PDF, first turn its pages into images with PDF to JPG, then run them through here.