OCR — recognize text in a PDF

A scanned contract, a photographed receipt, an old book — you get the same document, but searchable and copyable.

Drag a PDF or images here or

PDF, JPG, PNG · they stay on your device

    Sent to a server: 0 B

    FAQ

    Does OCR work in Serbian?

    Yes — Latin and Cyrillic Serbian, plus English. Pick the document language for the best result.

    Is the document sent to a server?

    No. Recognition is done by Tesseract (WebAssembly) in your browser. The first time, a language model (2–5 MB) is downloaded; after that it works offline too.

    Does the document look different afterwards?

    No. The original pages stay the same and an invisible text layer is added underneath — so search and copy work, but the document looks identical.

    Guides

    How to make a scanned PDF searchable To a computer, a scanned document is just a picture: it can't be searched, copied or read aloud. OCR fixes that.

    More tools

    Compress PDF

    Make a PDF smaller for email or uploading to a portal.

    Redact PDF

    Permanently remove ID numbers, account numbers or names — not just cover them.

    Merge PDF

    Combine several PDF files into one, in the order you choose.