I'm building generative music (non-AI) that lives on the web. Not just to play the music, to distribute it too. This the subject of the dissertation of my master's in Computer Music and Sound Design @ FCTUC, UC.
Why?
Media advances change playback
->
The web has unique playback features
->
Streaming platforms ignore it
->
We should explore them instead
Music media has always added playback features. Vinyl to CD added precise time-skipping, next and previous track controls, embedded metadata, and furthered the use of hidden tracks.
The web is an unique medium. It executes code. Accessible in any device with a browser (which is almost all modern devices).
Streaming platforms (like Spotify) ignore this. They just distribute static files. Yet generative music has precedent, even in the mainstream.
Conclusion -> explore how the web can run and distribute algorithmic music. See how building art for this medium changes the process, and its output.
Project
The platform's design will follow expected interactions, or break them, whichever serves the artwork.
The music is semi-stochastic. Seeds make each playthrough replayable, but never identical. Macro parameters like project duration shift between plays. So do small things like a rhythmic fill, a melodic turn.
Demo
Try an early demo here. It's my first test of making music with the tools I've built. I'm not contempt with the result, I think the aesthetic of the piece is not matching its environment. The final artwork will be different. The platform itself is working well.
░░░