Image to Text (OCR)
Pull the text out of a photo, screenshot or scan — right in your browser, in several languages.
Nothing is uploaded — this runs on your own device. Which is why the speed is your computer’s: a faster processor and more memory finish sooner, and a large file on a modest machine can take a few minutes. Please leave this tab open while it works.
How to use the Image to Text (OCR)
- Choose an image (a photo, screenshot or scan of text).
- Pick the language the text is written in.
- Click Extract text, then copy or download the result.
About the Image to Text (OCR)
This image to text tool uses OCR (optical character recognition) to read the words inside a picture and turn them into real, editable text. Point it at a screenshot, a photo of a document, a scanned page or a snap of a sign, and it gives you back text you can copy, search and paste — no retyping. It recognises several languages, including English, Spanish, French, Portuguese, German and Chinese, so you can pick whichever matches the text.
What makes it unusual is that everything runs in your browser. The OCR engine (a WebAssembly build of the well-known Tesseract project) and its language data are downloaded to your device the first time you run it, and the recognition happens locally — your image is never uploaded to a server. That means no sign-up, no watermark, and genuine privacy for sensitive documents.
For the best results, use a clear, well-lit image where the text is reasonably straight and in focus; faint, skewed or very low-resolution text is harder to read, just as it would be for a person. Because the engine loads on demand, the first run takes a few seconds to fetch the language pack, then subsequent runs are quicker. Once you have the text, you can tidy it in the online notepad or hear it read back with text to speech.
Frequently asked questions
Is my image uploaded to a server?
No. The OCR runs entirely in your browser using WebAssembly, so your image and the extracted text never leave your device.
Why does the first run take a moment?
The first time, your browser downloads the OCR engine and the language data for the language you picked. After that it is cached, so later runs are faster.
How can I get more accurate results?
Use a sharp, well-lit image with the text upright and in focus, and select the correct language. Clean screenshots and scans work best; blurry or angled photos are harder.
Which languages are supported?
English, Spanish, French, Portuguese, German and Simplified Chinese. Choose the one that matches the text in your image for the best accuracy.