Can anybody explain using the Nyquist-Shannon theorem why audio signals are usually sampled at 48,000 Hz? Hint: The normal range of human hearing is around 20 to 20,000 Hz Link to Wikipedia
Since the max frequency human can hear is 20kHz, the Audio signal needs to be sampled at twice the max frequency in order to prevent aliasing (according to Nyquist-Shannon theorem). So any sampling rate larger than about 40kHz would do the trick. But 48kHz is preferable since it is a multiple of the other standard sampling rates. The reason why we want them to be multiples of each other is to simplify the sampling rate conversion.