Clanker Me Not
Neural Network Bending in the DAW
What happens when you take a hammer to a neural network? Just as experimental musicians hacked toy keyboards to birth circuit bending, the same concept can be applied to audio generating neural networks. This talk explores the technical integration of neural audio network bending directly inside the DAW environment, moving past the pristine black box towards deliberate algorithmic malfunction.
We will discuss the structural architecture required to intercept, manipulate, and mutate the internal layers, weights, and latent paths of machine learning models mid-inference within a plugin framework. Looking under the hood of experimental VST/AU prototypes utilizing real-time neural audio and symbolic generation architectures, we will address the computational hurdles of low-latency model interaction. Attendees will leave with practical engineering strategies to turn rigid generative models into volatile, highly expressive musical instruments.
Leonardo Foletto
Leonardo Foletto is an independent audio software developer, live coding musician, and hardware hacker. His past work encompasses a wide spectrum of audio and music applications, from Digital Audio Workstations and audio plugins to music machine learning, automotive audio, and multimedia performance software.