How it works
- Drop a scanned PDF, or photos of a document. Several files are combined into one PDF.
- Click Recognise text. Each page takes a few seconds.
- Download a searchable PDF (you can search, select and copy its text), or copy the plain text.
Questions
Which languages are supported?
English for now. More languages are coming. Text in other Latin-alphabet languages is often recognised too, but accents may be missed.
How accurate is it?
Very accurate on clear printed text. Blurry photos, unusual fonts and handwriting give poorer results, so check names and numbers.
Why does the first run take longer?
The first time, your browser downloads the text recognition engine (about 7 MB). After that it's stored on your device and starts instantly, even offline.
Are my documents uploaded for OCR?
No. Recognition runs inside your browser. Your scans never leave your device.
Are my files uploaded?
No. SafeMerge runs entirely in your browser. Your files are read and changed on your own device and are never sent to any server, including ours.
Does it work offline?
Yes. Once the page has loaded you can turn off Wi-Fi and it keeps working. In Chrome or Edge you can also install SafeMerge as an app from the address bar.
Is it free?
Yes, with no sign-up, watermarks or file limits. Ads and donations pay for the site. Very large files are limited only by your device's memory.