What happens to sounds above the Nyquist frequency in digital audio? Do they get aliased into some type of noise?

We provide latest MatLab Homework Help, MatLab Assignment Help for students, engineers and researchers in Multiple Branches like ECE, EEE, CSE, Mechanical, Civil with 100% output.
Matlab Code for B.E, B.Tech,M.E,M.Tech, Ph.D. Scholars with 100% privacy guaranteed. Get MATLAB projects with source code for your learning and research.
In a properly designed system, there are none. They have to be entirely eliminated prior to sampling, otherwise you’re in trouble.

But I think you’re asking what happens if you do not do this filtering, or the filtering is inadequate.

This diagram shows you what happens in terms of a simple sine wave.

The green lines are where the signal (in red) is sampled. The red signal is above the Nyquist frequency, so the samples appear to trace out a different sinewave, the one in blue. As far as the digital side of things is concerned, these are just perfectly good samples, and when converted back to analogue, will produce the blue signal, a signal that was simply not present in the original signal.

The frequency of the blue signal is equal to the difference between the sampling frequency and the original signal, and so is not related in any harmonic way to the original. The blue signal is said to be an alias of the red signal — it’s the red signal in disguise.

Given a complex signal such as music, all the frequencies beyond the Nyquist will produce bogus low frequency aliases of the original signal. They are not related in any musical way, so it sounds appalling. Furthermore, because they’re just normal samples, you can’t get rid of them once they’re in the system. That’s why it is vital to remove them with a filter before there’s any chance that they can get sampled and cause havoc.

Incidentally, there is an everyday effect that demonstrates this very well. Watch an old western where you can see a wagon wheel rotating. Sometimes it seems to be going slow and even backwards. That’s exactly the same thing. The movie camera “sampled” the scene at 24 frames every second, but the spokes of the wheel are moving faster than that — above the Nyquist frequency. So the image you see is really a low frequency “alias” of the true scene, and so the wheel seems to revolve strangely. We can tolerate that fairly well for visual images, but for audio, all those incorrect frequencies are intolerable.