OCR for PDF scans

A scanned document is just pictures: you cannot select or search its text. OCR recognises the text and lays it as an invisible layer over the original pages, so the scan looks the same but the text can be copied and searched. Recognition runs in your browser, the file never reaches a server.