Neurorack - The first deep AI-based Eurorack audio synthesizer

We are thrilled to announce the birth of the Neurorack, the first deep AI-based real-time audio synthesizer, which comes in many formats and more specifically in the Eurorack format. The current prototype relies on the Jetson Nano for real-time synthesis. The model is based on a modified Neural-Source Filter (NSF) architecture, which allows real-time descriptor-based synthesis of percussive sounds. The goal of this project is to design the next generation of music instrument, providing a new tool for musician while enhancing the musician’s creativity. It proposes a novel approach to think and compose music. We deeply think that AI can be used to achieve this quest. The Eurorack hardware and software have been developed by the ACIDS team, with equal contributions from Ninon Devis, Philippe Esling and Martin Vert.

The project and design are fully open-source here: GitHub - ninon-io/Impact-Synth-Hardware

More details and papers on the science behind: Neurorack – Acids

4 Likes

Very nice! I like it.

I am only half joking when I say the next step would be to create a “human-like” robot (same skeletal movements) which learns to dance to that music.

2 Likes

Ahah thanks we’ll consider investing in a club for these :p

1 Like