In this video I'm showcasing a RAVE model that has been trained on a dataset consisting of Martsman material pre 2010 (subjectively heterogenous) using the V2 architecture and ELBo regularization.
The input is a stand alone break beat and I'm manipulating the latent vectors with different techniques. Model loading is done via nn~ object in Pure Data.

RAVE, is "A variational autoencoder for fast and high-quality neural audio synthesis” created by Antoine Caillon and Philippe Esling of Artificial Creative Intelligence and Data Science (ACIDS) at IRCAM, Paris. [ Ссылка ]
nn~ on GitHub: [ Ссылка ]

свернуть