Image to Text (OCR)
Runs in your browser. Your files never leave your device.
How we keep your files private
How to use Image to Text (OCR)
- Add a clear photo or scan of text.
- Choose the language of the text.
- Copy or download the recognised text.
Common uses
- Copying text from screenshots.
- Digitising printed pages.
- Extracting text from photos of signs or documents.
About this tool
Recognise printed text in images using Tesseract OCR compiled to WebAssembly. The recogniser and the language data download once and are cached. Everything runs on your device.
Tips and limits. Accuracy depends on image quality: sharp, well-lit, straight text with good contrast works best. Handwriting is not supported well.
Worked example
A clear photo of a printed page at about 300 dpi gives the best text. The tool reads the image with Tesseract in your browser and returns plain text you can copy.
Common problems and fixes
- The text has many mistakes.
- Use a sharper, well-lit photo, straight on, with the right language selected.
- Columns get mixed up.
- OCR reads left to right across the page. Crop to one column at a time.
- Handwriting is not recognised.
- OCR works for printed text. Handwriting needs specialised tools.
Next steps
Frequently asked questions
Which languages are supported?
English, Spanish, Portuguese, Hindi, Indonesian, Arabic, Turkish, Vietnamese, Russian, German, French and Japanese.
Why are there mistakes?
Blurry, skewed or low-contrast images reduce accuracy. Try a sharper photo.
Is my image uploaded?
No. Recognition runs in your browser with Tesseract.
Can it read handwriting?
Not reliably.
Are my files uploaded to a server?
No. Everything runs inside your browser on your own device, so your files and text are never sent anywhere.
Is it really free?
Yes. There is no sign-up, no watermark and no limit on how often you use it.
Does it work on my phone?
Yes. It works in any modern browser on phones, tablets and computers. Very large files may be slow on older phones.