Towards transferable speech emotion representation: on loss functions for cross-lingual latent representations
Publikation: Bidrag til bog/antologi/rapport › Konferencebidrag i proceedings › Forskning › fagfællebedømt
Dokumenter
- Fulltext
Indsendt manuskript, 798 KB, PDF-dokument
In recent years, speech emotion recognition (SER) has been used in wide ranging applications, from healthcare to the commercial sector. In addition to signal processing approaches, methods for SER now also use deep learning techniques which provide transfer learning possibilities. However, generalizing over languages, corpora and recording conditions is still an open challenge. In this work we address this gap by exploring loss functions that aid in transferability, specifically to non-tonal languages. We propose a variational autoencoder (VAE) with KL annealing and a semi-supervised VAE to obtain more consistent latent embedding distributions across data sets. To ensure transferability, the distribution of the latent embedding should be similar across non-tonal languages (data sets). We start by presenting a low-complexity SER based on a denoising-autoencoder, which achieves an unweighted classification accuracy of over 52.09% for four-class emotion classification. This performance is comparable to that of similar baseline methods. Following this, we employ a VAE, the semi-supervised VAE and the VAE with KL annealing to obtain a more regularized latent space. We show that while the DAE has the highest classification accuracy among the methods, the semi-supervised VAE has a comparable classification accuracy and a more consistent latent embedding distribution over data sets.
Originalsprog | Engelsk |
---|---|
Titel | ICASSP 2022 - 2022 IEEE International Conference on Acoustics, Speech and Signal Processing |
Forlag | IEEE |
Publikationsdato | 2022 |
Sider | 6452-6456 |
ISBN (Elektronisk) | 9781665405409 |
DOI | |
Status | Udgivet - 2022 |
Begivenhed | 47th IEEE International Conference on Acoustics, Speech, and Signal Processing, ICASSP 2022 - Virtual, Online, Singapore Varighed: 23 maj 2022 → 27 maj 2022 |
Konference
Konference | 47th IEEE International Conference on Acoustics, Speech, and Signal Processing, ICASSP 2022 |
---|---|
Land | Singapore |
By | Virtual, Online |
Periode | 23/05/2022 → 27/05/2022 |
Sponsor | Chinese and Oriental Languages Information Processing Society (COLPIS), Singapore Exhibition and Convention Bureau, The Chinese University of Hong Kong, Shenzhen (CUHK-Shenzhen), The Institute of Electrical and Electronics Engineers Signal Processing Society |
Navn | ICASSP, IEEE International Conference on Acoustics, Speech and Signal Processing - Proceedings |
---|---|
Vol/bind | 2022-May |
ISSN | 1520-6149 |
Bibliografisk note
Publisher Copyright:
© 2022 IEEE
ID: 324664969