Continuous descriptor-based control for deep audio synthesis

1 code implementation • 27 Feb 2023 • Ninon Devis, Nils Demerlé, Sarah Nabi, David Genova, Philippe Esling

Despite significant advances in deep models for music generation, the use of these techniques remains restricted to expert users.

Audio Synthesis Continuous Control +1

Creativity in the era of artificial intelligence

no code implementations • 13 Aug 2020 • Philippe Esling, Ninon Devis

As creativity is a highly context-prone concept, we underline the limits and deficiencies of current AI, requiring to move towards artificial creativity.

Ultra-light deep MIR by trimming lottery tickets

1 code implementation • 31 Jul 2020 • Philippe Esling, Theis Bazin, Adrien Bitton, Tristan Carsault, Ninon Devis

We show that our proposal can remove up to 90% of the model parameters without loss of accuracy, leading to ultra-light deep MIR models.

Audio Classification Drum Transcription +3

Diet deep generative audio models with structured lottery

1 code implementation • 31 Jul 2020 • Philippe Esling, Ninon Devis, Adrien Bitton, Antoine Caillon, Axel Chemla--Romeu-Santos, Constance Douwes

This hypothesis states that extremely efficient small sub-networks exist in deep models and would provide higher accuracy than larger models if trained in isolation.