About this tool
Read the text out of any image — a photo, screenshot, or scanned page — using the Tesseract OCR engine compiled to WebAssembly and running entirely in your browser. The image is never uploaded; the OCR model is downloaded once (then cached) and all recognition happens on your device.
Frequently asked questions
Is my image uploaded to an OCR server?
No. OCR normally runs in the cloud, but here the full Tesseract engine runs locally as WebAssembly — the image and the recognized text never leave your browser.
Why does the first run take a few seconds?
It loads the WASM OCR engine and the English language model (about 6 MB) the first time. Both are cached afterwards, so later runs start quickly.
How can I get the best results?
Use a clear, high-contrast, reasonably straight image. Recognition currently uses the English model and works best on printed text.