AudioStellar, an open source corpus-based musical instrument for latent sound structure discovery and sonic experimentation

GARBER, LEANDRO; CICCOLA, TOMÁS; AMUSATEGUI, JUAN CRUZ

Santiago

Conferencia; ICMC 2020; 2021

Generating a visual representation of short audio clips´ similarities is not only useful for organizing and exploring an audio sample library but it also opens up a new range of possibilities for sonic experimentation. We present Au-dioStellar, an open source software that enables creative practitioners to create AI generated 2D visualizations of their own audio corpus without programming or machine learning knowledge. Sound artists can play their input corpus by interacting with learned latent space using an user interface that provides built-in modes to experiment with. AudioStellar can interact with other software by MIDI sync-ing, sequencing, adding audio effects, and more. Creating novel forms of interaction is encouraged through OSC communication or writing custom C++ code using provided framework. AudioStellar has also proved useful as an educational strategy in courses and workshops for teaching concepts of programming, digital audio, machine learning and networks to young students in the digital art field.