Make a scanned PDF searchable
Reads the text in your scan and puts an invisible, selectable text layer behind the image — so you can search and copy from it, while it still looks identical.
- Nothing is uploaded
- No signup
- No watermark
- Works offline after the first visit
Questions
Does the OCR run on a server?
No. The recognition engine is compiled to WebAssembly and runs in your browser, and the language data is served from this site rather than from anyone else’s. It is fetched once, then cached, so later runs work with no network at all.
Why is it slow on long documents?
OCR is genuinely heavy work, and doing it on your device means your laptop is doing it rather than a datacentre. It runs in the background and shows per-page progress.