Pull the text out of a photo, scan or screenshot, and copy or save it.
Works best on clear, printed text. Handwriting, blurry photos and low contrast are read poorly.
Drop a photo or screenshot, or paste one with Ctrl+V.
Choose the language or languages of the text in the picture.
Edit the text if needed, then copy it or save it as a .txt file.
No. The text is read in your browser. Only the language data, a few MB per language, is downloaded the first time, and your browser remembers it afterwards.
More than 45 languages, including English, Hindi, Urdu, Arabic, Persian, Bengali, Punjabi, Tamil, Telugu, Chinese (Simplified and Traditional), Japanese, Korean, Thai, Spanish, French, German, Portuguese, Russian and Turkish. Search the list, and pick up to three at once for mixed text.
Very good on clear, printed text with good contrast, especially English, Arabic, Cyrillic, Devanagari (Hindi) and Chinese. Accents (such as French), and complex scripts such as Tamil or Telugu, can have mistakes, so check the result. Blurry photos, curved pages, small text and handwriting are read poorly. The confidence number shows how sure the reader is.
The reader and the language data need to download once. Later runs start much faster.
Not yet. Save the page as an image first, or use PDF to Image when it becomes available.
Use a sharp, straight, well-lit image, crop away everything except the text, and select every language that appears in it.