Image to text
Copy the text out of screenshots, photos and scans, right in your browser.
How it works
- 01Pick the language of the text, then press "Download & start" once (a few MB for the language).
- 02Drop in images or paste a screenshot with ⌘V / Ctrl+V. Add as many as you like.
- 03Check the highlighted words (yellow means unsure), edit the text if needed, then copy it or download a .txt.
Questions
›Which OCR model is used, and what is its license?
Tesseract, the open-source OCR engine started at HP and developed by Google and its community, with its "fast" neural-network language models (tessdata_fast). Both are under the Apache-2.0 license. It runs in your browser with WebAssembly on your processor: each language is a one-time download of 0.5–6 MB, plus about 4 MB for the engine, and is then reused from your browser.
›How accurate is it?
Very good on clear printed text: screenshots, documents and scans. Accuracy drops on handwriting, stylized fonts, text on busy backgrounds, curved pages and blurry or tilted photos. It's AI, so it can make mistakes: words it is unsure of are highlighted in yellow, and you can fix the text before copying it.
›Can it read several languages at once?
Yes. Pick up to three languages, for example English and German for a bilingual letter. Pick only the languages that are really in the image: each extra one makes reading a little slower and can add odd guesses.
›Does it keep line breaks and paragraphs?
Yes. The text comes out line by line with a blank line between paragraphs, as laid out in the image. Press "Join lines" to turn each paragraph into one line (mending words split with a hyphen) before pasting it into a document or email.
›Are my files uploaded anywhere?
No. Your files are processed on your own device by your browser. They never touch our servers, so there are no upload waits, no size caps from us, and nothing to delete afterwards.