Photo, screenshot or scan: get the text in seconds. The image stays on your device.
Optical character recognition (OCR) finds the letters in an image and turns them into editable text. The engine used, Tesseract, runs directly in your browser: the image is not sent anywhere.
.txt file.| Tip | Why |
|---|---|
| Sharp, well-lit photo | Blur is the first cause of errors |
| Document held straight | Slanted lines are split badly |
| Large enough text | Below about 10 pixels per letter, reading becomes unreliable |
| Correct language chosen | The engine uses the language's vocabulary to settle doubts |
Handwriting is poorly recognised, as is text on a very busy background. Layout (columns, tables) is not reproduced: you get plain text, to check before use.
For a PDF that already contains text, the "PDF to text" tool is faster and more reliable.
No. Recognition happens in your browser. Only the language model is downloaded the first time; your image never leaves your device.
The site's twelve languages: English, French, Spanish, German, Portuguese, Italian, Dutch, Polish, Turkish, Japanese, Russian and Arabic.
Poorly. It is designed for printed text; very neat handwriting may work, but expect mistakes.