Image to Text (OCR)
Copy the text out of screenshots, scanned pages and photos of documents. Reads 27 languages, including Hindi, Tamil, Bengali, Arabic, Chinese and Japanese, right in your browser.
Files stay on your device
How it works
- 1Add your imagesDrop screenshots, scans or photos of text, up to 50 at once.
- 2Pick the languageChoose the language the text is written in, then click Read text.
- 3Copy or downloadCopy each result, or download them as .txt files.
Questions
Are my images uploaded?
No. The tool runs inside your browser, so your files never leave your device and nobody else can see them.
How accurate is it?
Clean screenshots and scans are usually read almost perfectly. Photos work best when the text is straight, sharp and well lit. Each result shows how sure the reader was. Handwriting is rarely read well.
Which languages can it read?
English, Arabic, Bengali, Chinese (Simplified), Dutch, French, German, Gujarati, Hindi, Indonesian, Italian, Japanese, Kannada, Korean, Malayalam, Marathi, Polish, Portuguese, Punjabi, Russian, Spanish, Tamil, Telugu, Turkish, Ukrainian, Urdu and Vietnamese. If your text mixes in English words, like many forms and signs, turn on 'The text also has English in it'.
Why does the first run take longer?
The text reader and its language data (1 to 3 MB) are downloaded once, then kept by your browser. Later runs start straight away, even offline.
Is there a size limit?
Up to 50 images at once and 50 MB per file. Very large images (over 100 megapixels, or 60 on iPhone and iPad) are refused with a clear message so your browser doesn't run out of memory.