Skip to content

Image to Text

Extract the text from a photo, screenshot or scan. It is read by your own browser, so the picture is never uploaded. Copy the result or save it as a .txt file.

Common questions

  • No. The recognition engine is downloaded to your browser from this site and reads the image there. The picture and the text it contains are never sent anywhere. The first run downloads about 5 MB of engine and English language data, which your browser then reuses.

Getting the best result

Recognition works on the pixels it is given, so the picture matters more than any setting. Use a sharp image of printed text that is upright and evenly lit, ideally dark letters on a plain light background. Crop away everything that is not text. A screenshot or a flatbed scan will usually read almost perfectly; a photo taken at an angle, with glare or a patterned background, will not.

Private by design

Screenshots and photos of documents often contain things you would not paste into a stranger's website. The engine is loaded from this site and runs on your device, so the image and the text stay with you. The first run downloads about 5 MB (cached afterwards).

Limits

English only for now, and printed text only: handwriting is not read reliably. For a whole scanned document, use OCR PDF to keep the pages as they are and make them searchable instead of extracting text.