As all other works of the FORMS series I’ve been sharing here, this one is also about image sonification.
Here, I draw generative spectrograms using Processing, and sonify the resulting PNGs using an additive synthesizer in Max/MSP (a bank of 1080 sinusoidal waves). A counterpart of just using this additive synthesis method is that because the last partial is at 13289hz (MIDI note 127), there’s nothing else on the spectrum above that frequency, hence the gameboyish/low SR sound.
All the sounds (but one layer of the kick drum) are product of this image synthesis. A lot on research on sound synthesis to come up with bass, snares, bells, and the myriad of possible timbres you can sculpt when you understand a little bit about Fourier, harmonic series and acoustics, and start to pile partials (horizontal lines) to come up with cool sounds. The main mantra: create images which sound good.
Harmony, rhythm and effects (except for the mastering reverb) are also encoded in the image. I have implemented lots of generative algorithms (probabilistic sequencing, euclidean beats, multiperiodic rhythms, beat-repeat, arpeggiation, random generation of chords, melodies, dynamic contours, perlin noise…) to have ever-changing but controlled randomness, which brings the complex mashup character to the resulting music.
On this video, an improvisation i did today by making some preset jumps here and there. Scroll to check the different parts if you don’t want to get hipnotised for 30 minutes with generative complexity…
There’s still a lot of work to have the different behaviors “know” each other and make smart transitions. But still, i wanted to share the exercise to show the possibilities of the system.
Looking forward to have this ready for a live audiovisual gig!