Photo Text Reader (OCR)
Turn the text in photos and screenshots (Korean and English) into text you can copy. Photos are never uploaded; everything runs in this browser.
Add photos, screenshots or scans with clear text and it will read them. You can edit the result.
The first time, language data (about 2–5MB) is downloaded. After that the copy saved on this device is used.
Send this text to another tool
The text is not put in the address; it moves only inside this browser. Nothing is sent to a server.
For better results
- Large, sharp text reads best. Blurry, tilted or shadowed photos are less accurate.
- Handwriting, decorative fonts and very small text are hard to read.
- Results can be wrong, so always check important details (numbers, names, amounts) against the original. Low confidence is flagged.
- Very large photos are scaled down to 3000px on the long side before reading.
FAQ
Are my photos uploaded?
No. Recognition runs in your browser with WebAssembly and this site has nowhere to receive photos. The engine and language data are also files of this site, so no request goes outside.
Is it slow only the first time?
Yes. The first run downloads the engine and language data; after that the data saved on this device is used. Large photos can still take several seconds to tens of seconds.
Does it support Japanese or Chinese?
Only Korean and English for now.