Fabian Ostermann and Tim Löhde will present BbMuse, the AI framework developed by Ostermann, in a musical performance:
Abstract:
"We present a 10-minute live performance for experimental electronic music in which a continuously evolving neural network joins two human performers to form an improvising trio. Rather than training a generative model on pre-existing works, we use neuroevolution loosely inspired by NEAT to grow a feed-forward neural graph in real time, with no formal optimization objective. Following a tree metaphor, the performers engage the network in two complementary roles: one "plants" its seed by controlling parameter initialization and mutation, while the other "shapes" its canopy by mapping outputted MIDI and control events to different synthesizers and their parameters. The evolving topology is rendered in the Godot Engine as a live, wind-blown tree, giving the audience a direct view of the organism’s growth. Built on the open-source BbMuse framework, our approach foregrounds lightweight, real-time AI that learns alongside the musician without storing data, treating the machine as a creative partner rather than a replacement."
The annual Conference on AI Music Creativity (AIMC) brings together an interdisciplinary community of researchers and practitioners working at the intersection of AI and music. This makes the conference an exciting event, building a bridge between data science and music research and covering topics such as performance systems, computational creativity, machine listening, robotics, and sonification.
The 2026 edition of the conference is hosted by the Audio Communication Group of Technische Universität Berlin and the State Institute for Music Research (Staatliches Institut für Musikforschung). The conference will take place in the immediate vicinity of the Musikinstrumenten-Museum (Musical Instrument Museum).


Fabian Ostermann and Tim Löhde will present BbMuse, the AI framework developed by Ostermann, in a musical performance:
Abstract:
"We present a 10-minute live performance for experimental electronic music in which a continuously evolving neural network joins two human performers to form an improvising trio. Rather than training a generative model on pre-existing works, we use neuroevolution loosely inspired by NEAT to grow a feed-forward neural graph in real time, with no formal optimization objective. Following a tree metaphor, the performers engage the network in two complementary roles: one "plants" its seed by controlling parameter initialization and mutation, while the other "shapes" its canopy by mapping outputted MIDI and control events to different synthesizers and their parameters. The evolving topology is rendered in the Godot Engine as a live, wind-blown tree, giving the audience a direct view of the organism’s growth. Built on the open-source BbMuse framework, our approach foregrounds lightweight, real-time AI that learns alongside the musician without storing data, treating the machine as a creative partner rather than a replacement."
The annual Conference on AI Music Creativity (AIMC) brings together an interdisciplinary community of researchers and practitioners working at the intersection of AI and music. This makes the conference an exciting event, building a bridge between data science and music research and covering topics such as performance systems, computational creativity, machine listening, robotics, and sonification.
The 2026 edition of the conference is hosted by the Audio Communication Group of Technische Universität Berlin and the State Institute for Music Research (Staatliches Institut für Musikforschung). The conference will take place in the immediate vicinity of the Musikinstrumenten-Museum (Musical Instrument Museum).