How to use the OCR tool

Drop an image — a screenshot, a photo of a document, a scanned receipt — or click to choose a file. Pick the language (English or Chinese), then click Recognize text. The text appears below, ready to copy or download as a .txt file.

How it works (and why it's private)

Recognition runs entirely in your browser using an OCR engine compiled to WebAssembly. The image is decoded and processed on your device — it is never uploaded to a server, which makes the tool safe for documents you'd rather not share with a third party.

Note: the first recognition downloads the language model (~10 MB) from a CDN; after that it is cached by the browser and later runs are offline.

Tips for better results

• Use a sharp, well-lit image; text should be reasonably large and straight.
• Screenshots of websites and photos of printed text usually work best.
• Handwritten text is unreliable — print/digital text is the sweet spot.

FAQ

Is it really free? Yes — free, unlimited, no account, no watermark. Can it read Chinese? Yes, select Chinese before recognizing. Is my image uploaded? No — processing is 100% local.