How it works
OCR (optical character recognition) uses a neural network to read the letters in an image and turn them into text you can copy and edit.
Privacy
The AI runs on your own device. Your image is never uploaded to a server. The first time, it downloads the language model (a few MB), so it takes a little longer.
Tips for better results
- Use a sharp, well-lit image.
- The text should be straight, without much tilt.
- It works better on printed text than on handwriting.
Frequently asked questions
Is it free?
Yes, no sign-up and no limits.
Is my image uploaded to the internet?
No, recognition happens in your browser.
Does it read handwriting?
It works best on printed text; handwriting often fails.
Why is the first run slow?
It downloads the language model to your browser.
Can I paste the text into Word?
Yes, copy the text and paste it.