BWSNet: Automatic Perceptual Assessment of Audio Signals
Audio and Speech Processing
2024-01-23 v2 Sound
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.
Cite
@article{arxiv.2309.02592,
title = {BWSNet: Automatic Perceptual Assessment of Audio Signals},
author = {Clément Le Moine Veillon and Victor Rosi and Pablo Arias Sarah and Léane Salais and Nicolas Obin},
journal= {arXiv preprint arXiv:2309.02592},
year = {2024}
}