Image to Text reads the words in a picture and gives them back as plain text you can edit. Add a photo, screenshot or scan, press Read text, and the recognized text appears next to the image with a confidence score. It uses optical character recognition (OCR) that runs inside your browser.
You can drop or select images (JPG, PNG, WebP, GIF, BMP or HEIC), and read several at once. Copy the text or save it as a .txt file. English, Spanish, French, German, Portuguese and Italian are supported.
Your images never leave your device: the work happens in your browser, with no upload and no account. It is free and adds no watermark.
JPG, PNG, WebP, GIF, BMP and HEIC photos work. Screenshots, photos of printed pages, signs and scanned documents give the best results. Handwriting and decorative fonts are read poorly.
English, Spanish, French, German, Portuguese and Italian. Pick the language the text is written in before you read it. Text in other languages or scripts comes out garbled. The data for each language downloads once, the first time you use it.
The confidence score is the OCR engine's own estimate, from 0 to 100. Sharp, straight, high-contrast text usually scores above 85. Blurry, small, tilted or low-contrast images score lower, so proofread the result and fix mistakes in the text box.
Yes. Add as many images as you like. You can read the text of all of them joined together, or switch to one image at a time to check and edit its text.
No. The OCR engine runs in your browser and your images never leave your device. The engine and the English data download once, the first time you read an image.