Image to Text (OCR)
Turn a photo, screenshot or scan into editable text. Choose the language, run OCR in your browser and copy the result.
How to extract text from an image
- Choose an image containing text, such as a screenshot, a photo of a page or a scan.
- Select the language of the text. Leave Enhance image on for photos and scans.
- Click Extract text, then copy or download the result and fix any small errors.
Features
Free OCR
Powered by tesseract.js, running in your browser.
19 languages
Latin, Cyrillic, Arabic, Hindi, Thai, Chinese, Japanese and Korean.
Image enhancement
Optional grayscale and contrast boost for better accuracy.
Private
Your images are processed on your device and never uploaded.
Frequently asked questions
How accurate is the OCR?
On clear, high-contrast text of at least 300 pixels per line of text, accuracy is usually above 95%. Blurry photos, handwriting, unusual fonts and low resolution reduce it, so always proofread the result.
Why does the first run take a while?
The OCR engine and the language data are downloaded from a CDN the first time, around 10 MB for most languages. They are cached by your browser, so later runs start quickly.
Can it read handwriting?
Not reliably. It is designed for printed text. Neat block handwriting sometimes works, but cursive generally does not.
Is my image uploaded?
No. The image is processed on your device. Only the OCR engine and language files are downloaded; your picture never leaves your browser.