BWSNET: Automatic Perceptual Assessment of Audio Signals - Institut de Recherche et Coordination Acoustique/Musique
Conference Papers Year : 2024

BWSNET: Automatic Perceptual Assessment of Audio Signals

Abstract

This paper introduces BWSNet, a model that can be trained from raw human judgements obtained through a Best-Worst scaling (BWS) experiment. It maps sound samples into an embedded space that represents the perception of a studied attribute. To this end, we propose a set of cost functions and constraints, interpreting trial-wise ordinal relations as distance comparisons in a metric learning task. We tested our proposal on data from two BWS studies investigating the perception of speech social attitudes and timbral qualities. For both datasets, our results show that the structure of the latent space is faithful to human judgements.
Fichier principal
Vignette du fichier
ICASSP_2024-BWSNet_LeMoine.pdf (2.84 Mo) Télécharger le fichier
Origin Files produced by the author(s)

Dates and versions

hal-04722037 , version 1 (07-10-2024)

Identifiers

Cite

Clément Le Moine Veillon, Victor Rosi, Pablo Arias Sarah, Léane Salais, Nicolas Obin. BWSNET: Automatic Perceptual Assessment of Audio Signals. International Conference on Acoustics, Speech and Signal Processing (ICASSP 2024), Apr 2024, Séoul, South Korea. pp.10416 - 10420, ⟨10.1109/icassp48485.2024.10447014⟩. ⟨hal-04722037⟩
27 View
10 Download

Altmetric

Share

More