Image to Text (OCR)

Extract text from images, photos and screenshots with OCR in 10 languages — runs in your browser, nothing uploaded.

How do I extract text from an image?

Choose one or more images, select the language and press Extract text. Optical character recognition (OCR) by the open-source Tesseract engine reads the letters in your browser and returns editable text with a confidence score.

For the best accuracy

Use a sharp, evenly lit image with the text straight (not at an angle).
Dark text on a light background works best; crop away busy backgrounds with the Image Cropper.
Pick the correct language — mixed English + Sinhala or Tamil documents have their own options.
Screenshots usually give 95%+ accuracy; handwriting is not reliably recognised.

The first run downloads the language data (1–3 MB) to your browser; after that your browser keeps it, so later runs start faster.

Frequently asked questions

Are my images uploaded?

No. Tesseract runs on your device, so documents stay private.

Which languages are supported?

English, Sinhala, Tamil, Hindi, Arabic, Spanish, French, German, Portuguese and Simplified Chinese.

Can it read a PDF?

Convert the PDF pages to images with PDF to Image first, or use PDF to Text if the PDF has a text layer.