🔤 Image to Text (OCR)
Turn screenshots of comments, slides, receipts or book pages into text you can edit. Recognition runs locally with Tesseract.
✅ Free · No sign-up
Choose an image, drag it here, or press Ctrl/⌘+V to paste a screenshot
PNG, JPG, WebP, BMP, GIF · up to 30 MB
PNG, JPG, WebP, BMP, GIF · up to 30 MB
Hold Ctrl/⌘ to pick several. Each language downloads once (2–15 MB).
Options
How to use the Image to Text (OCR)
- Add an image by choosing, dropping or pasting it.
- Pick the language(s) in the image.
- Press Extract text, then correct any mistakes in the text box and copy or download it.
Why creators use it
- Copy text from a screenshot of a tweet, comment or slide.
- Digitise handwritten-style notes printed in a clear font, receipts or book pages.
- Pull quotes out of thumbnails or memes for captions.
Good to know
- OCR works best on clear, high-contrast printed text. Handwriting, stylised fonts and very small text give poor results.
- Always proofread: OCR can confuse similar characters (0/O, 1/l).
- Language data loads from the jsDelivr CDN the first time you use each language.
Frequently asked questions
Is the image uploaded?
No. Tesseract runs in your browser using WebAssembly.
How accurate is it?
For clean screenshots of printed text, typically very accurate. The tool shows Tesseract’s own confidence score so you know when to double-check.
Can it read PDFs?
Use the PDF OCR Converter, which renders each page and runs OCR on it.