The Aesthetics of Disharmony: Harnessing Sounds and Images for Dynamic Soundscapes Generation
summary
An autonomous system named Solato generates music and images simultaneously using rule-based mechanics plus procedural content grids of 3D assets and musical building blocks to simulate an endless train-journey soundscape. An ablation study shows the rules prevent asset overlap and visual clutter. An online evaluation with 74 participants compares Solato to a biased random baseline and a magenta/nightcafe pipeline; Solato rates higher on music, visuals, and audiovisual coherence and is judged closer to human-made.
AI-written summary · about
related talks
16mDemo: Functional Sound Design
04mThe Sound Sketchpad: Expressively Combining Large and Diverse Audio Collections
19mGenerative music, playful visualizations and where to find them
30mGenerating Music From Emotion (and other experiments)
54mIntegral scherzo: how machines compose and sing
39mSonic Earth - An Algorithmic Performance Approach to Real-Time Environmental Sonification
35mProgramming What Cannot Be Programmed: Aesthetics and Narrative
20mThe rules are the art: creating with AI