How to use Image to Text (OCR)
- Under Languages in the image, tick only the languages in your image.
- Drop one or more images, or paste a screenshot with Ctrl+V.
- Click Recognize text and wait while the engine and language data load.
- Check the words boxed in red, edit the text if needed, then copy it or download a .txt file.
Supported formats
In: JPG, PNG, WEBP, BMP, TIFF, HEIC and other images, or the first page of a PDF; several at once. Out: editable text you can copy or save as .txt, with all images combined in ocr.txt.
Limits
Works best with sharp, straight, printed text and good contrast; handwriting and stylized fonts are often misread. The first run downloads the language data (1–3 MB per language), which the browser then keeps.
Privacy
Everything runs in your browser: your files are never uploaded to a server, and there is no sign-up.
Frequently asked questions
Does OCR work for Vietnamese?
Yes. Tick Tiếng Việt and the accents are recognized. On the Vietnamese page it starts with Vietnamese and English selected.
Which languages are supported?
Vietnamese, English, Simplified and Traditional Chinese, Japanese, Korean, Hindi and Russian. You can tick several, but each extra language makes recognition slower.
Can it read handwriting?
Not reliably. The engine (Tesseract) is made for printed text, so handwriting is often misread.
How can I get better results?
Use a sharp, straight image with good contrast, keep Enlarge small images on for screenshots, and try the One line or One block layout for short snippets.
Are my images uploaded?
No. Recognition runs on your device with Tesseract; only the engine and language data are downloaded from this site.