How to use the Image to Text (OCR)
- Drop an image that contains text.
- Choose the language of the text.
- Press Extract text, then copy it or download a .txt file.
How it works
Text is recognised with Tesseract.js, an open-source OCR engine compiled to WebAssembly that runs inside your browser. The engine and language data download once from a CDN and are then cached; the image itself is read on your device and never uploaded.
Worked examples
Screenshot you can't copy from
Drop a screenshot of text that won't let you select it, run OCR, and paste the words wherever you need them.
Scanned page
Turn a photographed book page or receipt into editable text.
Limitations and assumptions
- Accuracy depends on the image — clear, straight, high-contrast text works best; blurry or handwritten text is harder.
- Pick the correct language first for good results.
- The first run downloads the engine (about 2–4 MB), so it takes a few seconds; later runs are fast.
Frequently asked questions
Is my image uploaded?
No. The OCR runs in your browser, so the image stays on your device. Only the open-source engine and language files are fetched from a CDN.
Which languages are supported?
English, Spanish, French, German, Italian, Portuguese, Dutch, Arabic, Hindi, Urdu, Russian, Chinese, Japanese and Korean.
Can it read handwriting?
It's built for printed text. Neat handwriting sometimes works, but results vary.
Built and reviewed by Guru G
Calculation logic is covered by automated tests against known values. Last reviewed Nov 14, 2026. Found a problem? Let us know — see our editorial policy.