About this tool
Turn a picture of text into text you can copy, search, and edit: screenshots of error messages, photographed pages, scanned letters, whiteboards, receipts, memes. The recognition runs on your device using the Tesseract OCR engine compiled to WebAssembly, so the image never leaves your computer, which matters for documents and IDs.
The first run downloads the engine and the language pack for the language you choose (about 10 MB); after that it is cached and works offline. Small screenshots are automatically enlarged before recognition, since OCR likes text about 30 pixels tall. Clear, straight, high-contrast text gives the best results.
How to use it
- Drop your PNG, JPG, WebP or GIF files into the box above, or click to browse. Nothing is uploaded.
- Adjust the options if you want; the defaults suit most files.
- Click the button and wait a moment.
- Download the results individually or all at once as a ZIP.
Questions
How accurate is it?
On clean printed text and screenshots, very. Handwriting, decorative fonts, low-resolution photos, and skewed pages reduce accuracy; crop and straighten the image first for better results.
Can it read PDFs?
Use PDF to Text for PDFs with real text. For scanned PDFs, convert the pages to images with PDF to PNG and run them through this tool.
Why is the first run slow?
The OCR engine and language data are downloaded once and cached. Recognition itself takes a few seconds per image.
Is it free, are files uploaded, which browsers work: the general questions are answered once for the whole site.