Text from a photo, privately

Photo of a document, a screenshot or a PDF — get editable text in seconds. Recognition runs right in your browser: nothing is uploaded.

Languages in the image

Stays on your device

Pick only the languages that are really in the image — fewer languages means faster and more accurate.

0bytes of your image are uploaded. Tesseract (open source, Apache-2.0) runs in this tab; only the language data is downloaded once and cached.
16languages: English, Russian, Ukrainian, German, French, Spanish, Italian, Portuguese, Polish, Turkish, Indonesian, Vietnamese, Chinese, Japanese, Korean, Arabic.
300dpi or a sharp, straight, well-lit photo gives the best result. Screenshots work great. Handwriting is not supported.