OCR - Extract Text from Image
Drop an image for text extraction
Extract printed text in English, Hindi and other scripts
Max file size: 50 MB
OCR language(s)
Useful for mixed-language images, e.g. English + Hindi. Choose up to 3 together.
Text is extracted as Unicode, including Hindi/Devanagari and other supported scripts. Tesseract recognizes glyphs, not the source font styling.
Image preparation
Recommended for scans and phone photos
100%
100%
Small images are enlarged for recognition. Auto enhance boosts faded print; Original is available when preprocessing hurts detail.
Tips for better OCR
- • Use a sharp, well-lit image with the page upright
- • Choose the script; add a second language for mixed text
- • Printed text works best; handwriting recognition is limited
- • First use downloads language data and caches it in this browser
- • Your image is processed locally and is not uploaded