Musings on computer music, with some musical amusements

"AlgoCompSynth" is a name I made up, from "algorithmic composition" and "digital sound synthesis". I've had the domain for years but have mostly just kept renewing it until now. If it's easier for you to think of it as "computer music", be my guest.

As the name implies, I'm interested in music composed and performed by digital computers. Despite those technologies inexorably moving into the mainstream, I still consider it experimental music.

Although I've dabbled in computer music for decades, even before the Commodore 64, I've mostly been busy doing other things and have published very little. Early in 2021, on the 20th anniversary of the passing of Iannis Xenakis, I decided it was time to focus on computer music more or less full-time. If you're interested in what I have done previously, my SoundCloud is at https://soundcloud.com/znmeb/albums.

I was an undergraduate at the University of Illinois when both Harry Partch and Lejaren Hiller were on the faculty. Iannix Xenakis was a major influence. So were Alwin Nikolais, Henri Pousseur, Charles Dodge, and Morton Subotnick. The music theory of Professor William A. Sethares is also a major influence.

But there have been others, in the popular, symphonic, and electronic realms. What I mostly listen to these days: Beethoven, Sibelius, Shostakovich, Prokofiev, Khachaturian, Gershwin, Hovhaness, Wendy Carlos, Loreena McKennitt and contemporary composers on YouTube:

• Sarah Belle Reid

• HAINBACH

• Helene Vogelsinger

• Dolores Catherino

• Khyam Allami

• Sevish

• Ela Minus

• bad snacks

1. Reworking "When Harry Met Iannis (2009)": I'm happy with the score, although I have no plans to generate any more music of that type. Markov chains make boring music; cutting the track lengths short as I did is a cop-out.

   But the sound design badly needs reworking. It was done with SFront and was supposedly a sung vowel emulation. It sounds muddy and awful and nothing like sung vowels. I'll probably synthesize a Partch diamond marimba and a chromelodeon for the rework.

2. Explorations of the sonicLAB Fundamental synthesizer: I've got the Sensel Morph / Buchla Thunder overlay for exploration, but Fundamental also has microtonal capabilities (think Partch) and stochastic controls (think Xenakis).

3. Using GPUs for synthesis: This is pure research, which is the most fun I can possibly have. I've got fourfive of them:

   • An NVIDIAⓇ Jetson™ Nano,
   • An NVIDIA Jetson Xavier NX,
   • An NVIDIA Jetson AGX Xavier,
   • A laptop with an NVIDIA GTX 1050 Ti, and
   • A desktop with an NVIDIA RTX 3090.

   I'm currently thinking along the lines of Professor Stefan Bilbao's work, but may end up with something simpler, like inverse FFTs. Or even Real-time additive synthesis with one million sinusoids using a GPU.

   The Jetson project has it's own web presence; see edgyR-containers: Docker Images for NVIDIAⓇ Jetson™ R Developers.

4. Sea chanteys: Long ago, I used to sing somewhat-less-than-family-friendly songs in a pub in Annapolis, MD, accompanying myself badly on a tenor guitar.

   But this is 2021 - I still have the guitar, but most of the lyrics won't fly any more, so I'm curious what sort of lyrics one of these new-fangled generative natural language processing gizmos can come up with. The jury is still out on whether I'll sing them myself or force a computer to do it.