OCR — image to text
Upload a scanned page or photo and get back editable text. Recognition runs entirely on your device — your image never leaves the browser. Pick the script, then review and fix the result.
How it works
- Upload a scanned page or photo (or drag it onto the box). It is read locally — nothing is sent to a server.
- Choose the script: Latin for French/English, Arabic for Arabic, Persian or Urdu.
- Press “Extract text”. The recognizer finds the lines and reads them on your device.
- Review and fix the text in the editor, then copy it or download a .txt file.
Frequently asked questions
Is my image uploaded anywhere?
No. Recognition runs entirely in your browser using on-device models. Your image and the extracted text never leave your device.
Which languages are supported?
Latin-script languages such as French and English, and Arabic-script languages — Arabic, Persian and Urdu. Pick the matching script before extracting.
How accurate is it?
On clear printed scans it is very accurate for Latin and good for Arabic. Low-quality scans, handwriting and decorative calligraphy (nasta'liq) are harder, so review the result and edit where needed.
Can I edit the result?
Yes. The recognized text opens in an editor so you can correct any mistakes before copying or downloading it.