PDF OCR

A scan is a photograph of a document: there is no text in it, which is why searching finds nothing and copying gives nothing. This renders each page, reads it with the Tesseract engine running inside your browser, and lays the recognised words back over the page in an invisible layer. The picture you see is untouched; what changes is that the words underneath it can now be found and copied.

Recognition is good, not perfect - clean print in a supported language comes out well, a creased receipt or a handwritten note does not. The result tells you how confident the engine was. English, Arabic, and eight other languages are available; the language data (a few megabytes) is downloaded from a public CDN the first time you use a language, and the page images and the text itself never leave your device.