Image to Text (OCR)
Extract text from images
About the Image to Text (OCR)
Turn screenshots, photos of documents and scanned pages into editable text with optical character recognition (OCR). Recognition runs on your device using Tesseract, the open-source OCR engine, so your images are not uploaded. The language data is downloaded once from a CDN the first time you use a language.
- Choose the language of the text.
- Choose, drop or paste (Ctrl+V) an image.
- Wait a few seconds while the text is recognised.
- Edit, copy or download the extracted text.
How to get accurate text
Use a straight, sharp, well-lit image with dark text on a light background. Crop out everything except the text, and pick the right language before you start. Screenshots give the best results of all.
Frequently asked
Are my images uploaded?
No. OCR runs in your browser. Only the OCR engine and language files are downloaded, never your image.
How can I get better results?
Use a sharp, well-lit image with straight text and good contrast. Crop out everything except the text.
Can it read handwriting?
It works best on printed text. Neat handwriting may be partly recognised, but results vary.
Can it read Hindi and other Indian languages?
Yes: Hindi, Marathi, Bengali, Tamil, Telugu and Gujarati are included, plus English + Hindi together for mixed documents.
More from the Photo, Video & Audio drawer
Other tools kept in this drawer, plus a few favourites from the bench.