Image to Text (OCR)
100% in your browser — nothing is uploadedPrivacy note: recognition runs in your browser via WebAssembly. The OCR engine itself (a few MB) downloads from a CDN on first use; your images are never uploaded anywhere.
Retyping text from a screenshot or photographed page is a chore computers solved decades ago. Drop the image here and optical character recognition extracts the text into an editable box you can copy from. Unusually for an online OCR tool, recognition runs in your browser via WebAssembly (the Tesseract engine) — the image is never uploaded, which matters when it's a photographed contract or ID.
Accuracy is honest, not magical: crisp screenshots and clean scans read nearly perfectly; phone photos at angles, handwriting, and low-contrast text degrade results. Better light and a straighter shot beat any setting.
How to use the image to text (ocr)
- Drop a photo, screenshot, or scan containing text (JPG/PNG/WebP).
- Wait a few seconds — the engine downloads to your browser on first use, then recognizes locally.
- Edit the extracted text in place and copy it.
Frequently asked questions
Which languages are supported?
v1 recognizes English. The underlying engine supports 100+ languages and a language picker is on the roadmap.
Can it read handwriting?
Mostly no — Tesseract is built for printed text. Neat block capitals sometimes work; cursive rarely does.
Why was the first run slow?
The OCR engine (a few MB of WebAssembly and language data) downloads from a CDN once, then is cached. Recognition itself takes a few seconds per image after that.
Are my images uploaded for processing?
No — this is the rare OCR tool where recognition genuinely runs on your device. The engine downloads to the browser; the images stay with you.