Enrolment-based personalisation for improving individual-level fairness in speech emotion recognition

Fuente: Zenodo
Enregistré dans:
Détails bibliographiques
Auteurs principaux: Triantafyllopoulos, Andreas, Schuller, Björn
Format: Recurso digital
Langue:anglais
Publié: Zenodo 2024
Sujets:
Accès en ligne:
Tags: Ajouter un tag
Pas de tags, Soyez le premier à ajouter un tag!
_version_ 1866901174176710656
author Triantafyllopoulos, Andreas
Schuller, Björn
author_facet Triantafyllopoulos, Andreas
Schuller, Björn
contents <p>The expression of emotion is highly individualistic. However, contemporary speech emotion recognition (SER) systems typically rely on population-level models that adopt a `one-size-fits-all' approach for predicting emotion. Moreover, standard evaluation practices measure performance also on the population level, thus failing to characterise how models work across different speakers. In the present contribution, we present a new method for capitalising on individual differences to adapt an SER model to each new speaker using a minimal set of enrolment utterances. In addition, we present novel evaluation schemes for measuring fairness across different speakers. Our findings show that aggregated evaluation metrics may obfuscate fairness issues on the individual-level, which are uncovered by our evaluation, and that our proposed method can improve performance both in aggregated and disaggregated terms.</p>
format Recurso digital
id zenodo_https___doi_org_10_21437_interspeech_2024-98
institution Zenodo
language eng
publishDate 2024
publisher Zenodo
record_format zenodo
spellingShingle Enrolment-based personalisation for improving individual-level fairness in speech emotion recognition
Triantafyllopoulos, Andreas
Schuller, Björn
personalisation
fairness
speech emotion recognition
computational paralinguistics
deep learning
<p>The expression of emotion is highly individualistic. However, contemporary speech emotion recognition (SER) systems typically rely on population-level models that adopt a `one-size-fits-all' approach for predicting emotion. Moreover, standard evaluation practices measure performance also on the population level, thus failing to characterise how models work across different speakers. In the present contribution, we present a new method for capitalising on individual differences to adapt an SER model to each new speaker using a minimal set of enrolment utterances. In addition, we present novel evaluation schemes for measuring fairness across different speakers. Our findings show that aggregated evaluation metrics may obfuscate fairness issues on the individual-level, which are uncovered by our evaluation, and that our proposed method can improve performance both in aggregated and disaggregated terms.</p>
title Enrolment-based personalisation for improving individual-level fairness in speech emotion recognition
topic personalisation
fairness
speech emotion recognition
computational paralinguistics
deep learning
url https://doi.org/10.21437/interspeech.2024-98