I assume that they mean that they want the sample rate to be 48000 Hz (48 kHz), which gives an audio bandwidth of 24 kHz (half the sample rate).
Note that “32 bit” could be “signed 32-bit integer”, “unsigned 32-bit integer” or “32-bit floating point (IEEE)”.
I’ve never come across “unsigned 32-bit integer” audio, so I’d say that was very unlikely to be what they mean.
“32-bit float” (floating point) is what Audacity uses internally. They could mean this.
“Signed 32-bit integer” is used by some other apps. They could mean this.
In general, there are three common sampling rates: 44100 is the one for Audio CD and while it’s not a gift from the angels. it does OK and everybody knows what it is. 48000 is the sample rate for video. They got a later start and didn’t have to put up with the restrictions of the early CD disk.
Studios generally use 96000. As people point out constantly, you can’t hear that improvement, but it guarantees perfect multi-pass editing and you can tell the client their work is the same standard as Warner Brothers and Glen-Glenn Sound.
Either 48000 or 96000 sampling rate will give you audio transmission up to 24000Hz.
The odd duck is 32 bit format.
CD and Video use 16 bit and Studios use 24 bit. Audacity uses 32-bit Floating internally. Nobody breathing uses plain 32-bit.
But it’s not that hard to do. Make sure your system is at 48000 and it says that in the Audacity lower-left window.
Export > WAV > and 32-bit is one of the export options.
I told the Mac to give me INFO on the sound file and it did.
Note in the middle bunch: 1 channel (mono), 48000 - 32 bit.
The exact specification they wanted.
You still have to produce actual work of that quality. Are you doing this on the kitchen table? Good luck.