Nonnegative Tucker Decomposition with Beta-divergence for Music Structure Analysis of audio signals - Archive ouverte HAL Accéder directement au contenu
Communication Dans Un Congrès Année : 2022

Nonnegative Tucker Decomposition with Beta-divergence for Music Structure Analysis of audio signals

Résumé

Nonnegative Tucker Decomposition (NTD), a tensor decomposition model, has received increased interest in the recent years because of its ability to blindly extract meaningful patterns in tensor data. Nevertheless, existing algorithms to compute NTD are mostly designed for the Euclidean loss. On the other hand, NTD has recently proven to be a powerful tool in Music Information Retrieval. This work proposes a Multiplicative Updates algorithm to compute NTD with the beta-divergence loss, often considered a better loss for audio processing. We notably show how to implement efficiently the multiplicative rules using tensor algebra, a naive approach being intractable. Finally, we show on a Music Structure Analysis task that unsupervised NTD fitted with beta-divergence loss outperforms earlier results obtained with the Euclidean loss.

Dates et versions

hal-03409508 , version 1 (29-10-2021)

Licence

Paternité

Identifiants

Citer

Axel Marmoret, Florian Voorwinden, Valentin Leplat, Jérémy E Cohen, Frédéric Bimbot. Nonnegative Tucker Decomposition with Beta-divergence for Music Structure Analysis of audio signals. GRETSI, XXVIIIème Colloque Francophone de Traitement du Signal et des Images, Sep 2022, Nancy, France. ⟨hal-03409508⟩
136 Consultations
0 Téléchargements

Altmetric

Partager

Gmail Facebook X LinkedIn More