I've occasionally used CamScanner, and it came highly recommended to me by academic researchers who need to take photos of original documents for later perusal. However, I've never really made heavy use of it, mostly because my phone camera just doesn't take that great of photos. This is probably more like what you want.
The main thing I use is Canon imageFORMULA P-215 Scan-tini Personal Document Scanner. It does double-sided scanning, OCR has worked well for everything I've ever tried to use it for, and scans quickly. It was defs a more spendy purchase, but given that I can barely get the scanner on my printer to work for me 90% of the time, it was worth it to me. I bought it on the recommendation of The Wirecutter. It looks like they've since updated their recommendations for both the fancypants and cheap categories of standalone scanners.
Does the scanner integrate with CamScanner somehow? I mentionned OCR so it can recognize the document's contents and attach this as metadata to the image to make searching easier. I think what you are mentioning has more to do with scanning the document on your computer, extracting the text from it, discarding the image and keeping only the text; right?
Sorry, I should have been more clear.
CamScanner and the Canon scanner are completely separate things.
CamScanner theoretically has OCR if you get the licensed version, but I've never used it so I can't speak to how well it works. I mentioned it as a phone app that might meet your requirements.
When I use the Canon scanner, I just scan the documents in, save them as PDF, and am done with it. The PDFs are OCR'd, but my main use is for searching. I keep the 'original' scan pdf incase I need to print things (it's happened), and shred the paper hardcopy. I have a folder on my computer that is assorted documents I might one day need to look at again.