Filozy

Local PDF processing

OCR a scanned PDF without uploading it

Make a scanned PDF searchable on your own device. The text-recognition engine runs inside your browser, so records, IDs and contracts never reach an OCR service.

Open OCR PDF →
0document bytes sent

Offline-like, without installing an app.

OCR is where privacy usually breaks: the documents people scan are exactly the ones they should not upload. Filozy downloads the recognition engine and its language data from its own origin, runs them in a Web Worker on your device, and adds an invisible text layer over the original page images. You get a searchable PDF and a plain-text file, and the scan itself stays put.

How it works

Three transparent moments

  1. 1

    Choose the scan

    Select an image-only PDF. Recognition is the slowest job on this site, so split a very long scan into the part you need first.

  2. 2

    Let the browser read it

    Each page is rendered and recognised locally. Progress and a confidence figure are shown as it works.

  3. 3

    Save searchable output

    Download the searchable PDF, the extracted text, or send the result on to Redact, Compress or Protect without re-selecting it.