Fiche de CAILLON Antoine

Titre : Hierarchical temporal learning for music neural audio synthesis
Date : 21 février 2023
Heure: 09:00
Résumé : In this thesis, we study how a hierarchical approach to audio modeling can address the musical signal modeling task, while offering different levels of control to the user. Our main hypothesis is that extracting different representation levels of an audio signal allows to abstract the complexity of lower levels for each modeling stage. We start by addressing raw audio modeling by proposing an audio model combining a Variational AutoEncoder and an adversarial training strategy, yielding fast and high-quality 48kHz neural audio synthesis. We then study how autoregressive models can be used to model the temporal behavior of the representation yielded by this low-level audio model. Finally, we propose a set of techniques allowing the use of all previous methods in realtime applications that we developed during this thesis. We conclude by presenting various creative collaborations led in parallel of this work with several composers and musicians.
Lieu : Salle Igor Stravinsky, IRCAM, 1 Place Igor Stravinsky, 75004 Paris FRANCE