So Perlin Noise is a kind of noise (or order) that is capable of generating graphic scenes which resonate with either beauty or reality. My assumption is that it should follow that the same will happen with sound. I might be wrong.

9 orders of Perlin Noise |or| random data generated at 9 different frequencies. In SuperCollider

I’ve now got Perlin Noise being generated in SuperCollider. The 2 screenshots illustrate the random data generated in 9 different frequencies… and then all added up into 1 signal. That’s a one dimensional Perlin Noise signal (of 9th order). You can already see that the profile of the signal could soon start resembling a landscape.

9 orders added up into 1 dimensional Perlin Noise

I’ve then tried using that signal to affect audio in a number of different ways. So far, I have not found anything particularly interesting. I’ve used the Perlin Noise to:

1. drive the pitch of a Sine oscillator
2. drive the volume of a Sine oscillator
3. drive the position parameter of a Granular Synthesis voice.

None of these have yet given me results which bowl me over.

I have to ask myself the question: which part of a sound might be well described by this type of order. Perhaps it is FFTs. The above Perlin Noise screenshot certainly looks like an FFT. Lots more trying out to do.