Towards Conditional Adversarial Training for Predicting Emotions from Speech - Archive ouverte HAL Accéder directement au contenu
Communication Dans Un Congrès Année : 2018

Towards Conditional Adversarial Training for Predicting Emotions from Speech

Résumé

Motivated by the encouraging results recently obtained by genera-tive adversarial networks in various image processing tasks, we propose a conditional adversarial training framework to predict dimensional representations of emotion, i. e., arousal and valence, from speech signals. The framework consists of two networks, trained in an adversarial manner: The first network tries to predict emotion from acoustic features, while the second network aims at distinguishing between the predictions provided by the first network and the emotion labels from the database using the acoustic features as conditional information. We evaluate the performance of the proposed conditional adversarial training framework on the widely used emotion database RECOLA. Experimental results show that the proposed training strategy outperforms the conventional training method, and is comparable with, or even superior to other recently reported approaches, including deep and end-to-end learning.
Fichier principal
Vignette du fichier
Han18-TCA.pdf (262.16 Ko) Télécharger le fichier
Origine : Fichiers produits par l'(les) auteur(s)
Loading...

Dates et versions

hal-01994215 , version 1 (25-01-2019)

Identifiants

  • HAL Id : hal-01994215 , version 1

Citer

Jing Han, Zixing Zhang, Zhao Ren, Fabien Ringeval, Björn Schuller. Towards Conditional Adversarial Training for Predicting Emotions from Speech. ICASSP 2018 - 2018 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP), Apr 2018, Calgary, Canada. pp.6822-6826. ⟨hal-01994215⟩
45 Consultations
381 Téléchargements

Partager

Gmail Facebook X LinkedIn More