Reading Scanned PDFs (Text Recognition)

Not every PDF an applicant uploads has real text inside it. Bank statements, certificates, and signed letters are often scans or phone photos: a picture of a page, not the page itself. On a document like that you normally can't select a line, copy a figure, or search for a name, and the citations Optible shows you have nothing to anchor to.

Optible handles this for you. When you open a document, it reads any scanned pages so they behave like ordinary text. There's nothing to switch on.

What You'll Notice

The first time you open a scanned page, Optible takes a few seconds to set itself up, and you'll see a small progress bar at the top of the viewer while it does. This happens only once per session. After that, scanned pages are read quietly as you scroll, with no banner.

Once a page has been read, you can select and copy from it, search it, and see citations highlighted on it, just like any other document. If you click a citation that points to a scanned page, Optible reads that page right away so the highlight lands in the right spot.

Where Your Documents Go

Nowhere. The reading happens inside your own browser, on your device. The file isn't sent anywhere to be read. Once a page has been read, the result is kept in your browser, so it doesn't have to be read again if you close the document and come back to it later.

When It's Accurate, and When to Double-Check

Text reading is tuned for English, and it's only as good as the scan it's working from. A clean, straight scan reads accurately. A blurry photo, a page scanned at an angle, faint print, or handwriting can come through with mistakes or gaps, so check anything that matters against the original page in front of you.

Documents that already contain real text don't need any of this, and you won't see the setup step on them. One thing worth knowing if you work in a private or incognito window: Optible can't keep the read results between sessions there, so a scanned document may be read again each time you open it.