OCR
Image to Text (OCR)
Pull the text out of a screenshot, scan, or photo — copy it or download it as a plain text file.
Your files never leave your browser. Nothing is uploaded, stored, or seen by us.
How it works
This tool runs Tesseract.js, a WebAssembly build of the open-source Tesseract OCR engine, directly in your browser. When you click "Extract text," the engine and its language model load on demand and analyze your image locally — no server, no upload, no image data ever transmitted.
Frequently asked questions
- What is OCR?
- OCR (optical character recognition) is a technology that detects and extracts readable text from an image — turning a picture of words into actual, selectable, copyable text.
- What image types work best for OCR?
- Clear, high-contrast text works best — clean screenshots, scanned documents, and well-lit photos of printed text. Handwriting, low-resolution images, heavy stylized fonts, and text at extreme angles are harder to recognize accurately.
- Does this tool support languages other than English?
- The current version is configured for English text recognition. If you regularly need other languages, let us know which ones and we'll consider adding language packs.
- Is my image sent anywhere for text recognition?
- No. This tool uses Tesseract.js, an OCR engine compiled to run entirely in your browser via WebAssembly. The OCR model downloads once (a few megabytes) and then processes your image locally — nothing is uploaded.
- Why does the OCR engine take a moment to start?
- To keep every other page on this site fast, the OCR engine is only loaded when you click "Extract text" on this specific page — it's a genuinely large library, so we don't ship it anywhere it isn't needed.