Extract text from an image
OCR for screenshots and photos, English and Hindi.
Files stay on your device. No accounts. Free.
Loading tool...
About this tool
Drop a screenshot or a photo of a document and get its text, ready to copy or download. Recognition runs on Tesseract, the open source OCR engine, compiled to WebAssembly and executed inside your browser tab. English, Hindi, or both together.
One honest note: your image is never transmitted, but the recognition model itself (a few MB per language) is downloaded from a CDN the first time you use it. Accuracy is best on sharp, straight, high-contrast text; handwriting and low-light photos will be rough.
How it works
- Drop one or more images, they stay on your device.
- Pick your options and convert.
- Download the result; multiple files arrive as a single zip.
Frequently asked questions
Is my image uploaded for the OCR?
No. The Tesseract engine runs as WebAssembly in your browser and processes the image on your device. What does get downloaded, once, is the language model itself, a generic few-MB file that is the same for everyone.
How accurate is the text recognition?
On clean screenshots and scans it is very good. On photos it depends on lighting, focus and angle, and handwriting is mostly beyond it. Always give the output a quick read before using it.
Which languages are supported?
English and Hindi, individually or combined for mixed documents. Each language model downloads on first use and is cached by your browser after that.