Désentrelacement Fréquentiel Doux pour les Codecs Audio Neuronaux

Fuente: arXiv
Enregistré dans:
Détails bibliographiques
Auteurs principaux: Giniès, Benoît, Bie, Xiaoyu, Fercoq, Olivier, Richard, Gaël
Format: Preprint
Publié: 2025
Sujets:
Accès en ligne:
Tags: Ajouter un tag
Pas de tags, Soyez le premier à ajouter un tag!
_version_ 1866915533538983936
author Giniès, Benoît
Bie, Xiaoyu
Fercoq, Olivier
Richard, Gaël
author_facet Giniès, Benoît
Bie, Xiaoyu
Fercoq, Olivier
Richard, Gaël
contents While neural-based models have led to significant advancements in audio feature extraction, the interpretability of the learned representations remains a critical challenge. To address this, disentanglement techniques have been integrated into discrete neural audio codecs to impose structure on the extracted tokens. However, these approaches often exhibit strong dependencies on specific datasets or task formulations. In this work, we propose a disentangled neural audio codec that leverages spectral decomposition of time-domain signals to enhance representation interpretability. Experimental evaluations demonstrate that our method surpasses a state-of-the-art baseline in both reconstruction fidelity and perceptual quality.
format Preprint
id arxiv_https___arxiv_org_abs_2510_03741
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle Désentrelacement Fréquentiel Doux pour les Codecs Audio Neuronaux
Giniès, Benoît
Bie, Xiaoyu
Fercoq, Olivier
Richard, Gaël
Sound
Audio and Speech Processing
Neurons and Cognition
While neural-based models have led to significant advancements in audio feature extraction, the interpretability of the learned representations remains a critical challenge. To address this, disentanglement techniques have been integrated into discrete neural audio codecs to impose structure on the extracted tokens. However, these approaches often exhibit strong dependencies on specific datasets or task formulations. In this work, we propose a disentangled neural audio codec that leverages spectral decomposition of time-domain signals to enhance representation interpretability. Experimental evaluations demonstrate that our method surpasses a state-of-the-art baseline in both reconstruction fidelity and perceptual quality.
title Désentrelacement Fréquentiel Doux pour les Codecs Audio Neuronaux
topic Sound
Audio and Speech Processing
Neurons and Cognition
url https://arxiv.org/abs/2510.03741