It's a little more complicated.
The audio source was originally digital or analog, was transferred to digital-land/recorded at 44.1, 96, or even 192, and then was output as analog, and is now being re-recorded digitally. I think 192 recording is just a waste of space. But does it make sense to record at 96 as opposed to 44.1, given that it went through an (high quality) analog stage?
"CD Quality" (16-bit/44.1kHz is better than human hearing) so 44.1 or 48kHz is fine.
The only reason to record any higher would be to have more precision for cleanup, but you'd probably end up recording a lot more noise, too.
Thanks to all who responded.
If your ADC doesn't introduce noticeable aliasing from incoming ultrasound (AFAIK, only absolutely terrible ones do that) and if you don't plan to slow down the recording (thereby shifting ultrasound into audible range) then there's no benefit of higher-than-necessary sampling rate.
If your ADC doesn't introduce noticeable aliasing from incoming ultrasound (AFAIK, only absolutely terrible ones do that)
Well ... If it doesn't, what then? Because it puts on an analog filter starting somewhere, or because it internally uses higher precision and then writes to 44.1? Checking the technical specifications wouldn't hurt. Especially if it is "old" gear ...
On a different note: there are still DACs around that will make annoying noises when having to switch between 44.1k and 48k material or the other way around. Say, if this is to be played back in circumstances where there is not a single CD source but lots of 48k material, then going 48k consistently could save some minor nuissances. OTOH, if everything else in the library is CD rips ...