Tag Archives: audio theory

Snake Extension – What I have learned in the Spring semester

I taught Digital Signal Processing Theory class in the last Spring. I have never studied this hard to teach a course. As a result, I learned (or relearned) about DSP quite a bit. More specifically, I learned to appreciate elegance of “classic” techniques, such as AM, FM, sample-and-hold, and delays. I think I can explain them in digital terms. Most importantly, I learned that digital signal processing is all about (good) math.

The best way to wrap up the semester and summarize what I have learned is to make a piece using the new techniques I have learned.

1. To start the piece, I begin with a sinusoid generated with very slow Low Frequency Oscillator (LFO) on its pitch. The LFO’s shape is made with a  wave shaping function using polynomials. The result is an unusual curve pitch pattern.

2. On this polynomial curve, I add sample and hold. The smooth line gets “steps” in terms of pitch. I change the S&H rate to create different rhythms

3. As the piece progresses, I would like to have some frequency modulation gradually fading in.

4. This should sound fun if I have more of them. Here is an example of all the techniques with four sinusoids. The rate of S&H, the shape of polynomial curve, and the modulation rate of FM are randomly selected for each line.

The resulting sound sounded like a good accompaniment for a noisy electronic piece. So I played a no-input mixer and custom synthesizer over the polynomial pads. I also continued the piece with algorithmically drum part that I have developed for Snake and Ox track in my latest album (more about this track later).

Here’s the final result, Snake Extension. I think I’ll add this to my solo repertoire.