Runs in your browser · nothing is uploaded
Image to Text (OCR)
Pull the text out of screenshots, scans and photos with OCR that runs on your device — crop, rotate, pick a language, then edit, copy or download the result.
Getting the best result
- Use the highest-resolution image you have; text should be at least ~20 pixels tall.
- Rotate so the text reads left to right, upright.
- Crop away photos, logos and other columns.
- Pick the language the text is written in.
About the engine
Recognition uses tesseract.js (Apache-2.0), a WebAssembly build of the Tesseract OCR engine, with its LSTM “best (integer)” language models. If no text is found, the tool says so rather than guessing.
Frequently asked questions
How do I get text from an image?+
Add a screenshot, scan or photo, pick the language and press “Read text”. The recognised text appears in an editable box you can copy or save as a .txt file.
Is my image uploaded to a server?+
No. Text recognition runs on your device with the open-source Tesseract engine (tesseract.js) in a background worker. The engine and English language data are loaded from this site; other languages are downloaded from cdn.jsdelivr.net only if you agree — your images are never sent anywhere.
Why is some text wrong?+
OCR works best on sharp, straight, printed text with good contrast. Blurry photos, handwriting, decorative fonts, tables and multi-column pages often produce mistakes or text out of order. Crop to the text, rotate it upright and choose the right language to improve results — and always check the output.
Can it read handwriting?+
Not reliably. The engine is trained on printed text; neat handwriting sometimes works, but expect errors.
What does the confidence number mean?+
It's the engine's own average confidence in the words it found. Below about 60% the result is likely to contain mistakes. It's a guide, not a guarantee.
Can I process several images?+
Yes. Each image keeps its own result, which you can copy or download separately, or save all results in one .txt file.