Scanned contracts, invoices and letters are just pictures inside a PDF, so you can't search or copy them. This tool reads each page with Arabic and English text recognition and adds an invisible text layer, so the file looks exactly the same but its words can be found with Ctrl+F, selected and copied.
No. The original page images are kept untouched; only invisible text is added on top.
Clear printed Arabic and English usually read well. The result shows the average accuracy; blurry photos, handwriting and stamps are harder to read.
Each page is read on your own device, which keeps your document private. A typical page takes a few seconds; the first run also downloads the recognition data once.
Everything happens in your browser. Your files are never uploaded to a server, so nobody else can see them.