Dynamic Speech Emotion Recognition with State-Space Models

Abstract : Automatic emotion recognition from speech has been focused mainly on identifying categorical or static affect states, but the spectrum of human emotion is continuous and time-varying. In this paper, we present a recognition system for dynamic speech emotion based on state-space models (SSMs). The prediction of the unknown emotion trajectory in the affect space spanned by Arousal, Valence, and Dominance (A-V-D) descriptors is cast as a time series filtering task. The state- space models we investigated include a standard linear model (Kalman filter) as well as novel non-linear, non-parametric Gaussian Processes (GP) based SSM. We use the AVEC 2014 database for evaluation, which provides ground truth A-V-D labels which allows state and measurement functions to be learned separately simplifying the model training. For the filtering with GP SSM, we used two approximation methods: a recently proposed analytic method and Particle filter. All models were evaluated in terms of average Pearson correla- tion R and root mean square error (RMSE). The results show that using the same feature vectors, the GP SSMs achieve twice higher correlation and twice smaller RMSE than a Kalman filter.
Type de document :
Communication dans un congrès
23rd European Signal Processing Conference (EUSIPCO), Aug 2015, Nice, France. 2015
Liste complète des métadonnées

https://hal-imt.archives-ouvertes.fr/hal-01198424
Contributeur : François Septier <>
Soumis le : samedi 12 septembre 2015 - 15:59:57
Dernière modification le : jeudi 11 janvier 2018 - 02:08:23

Identifiants

  • HAL Id : hal-01198424, version 1

Citation

Konstantin Markov, Tomoko Matsui, François Septier, Gareth W. Peters. Dynamic Speech Emotion Recognition with State-Space Models. 23rd European Signal Processing Conference (EUSIPCO), Aug 2015, Nice, France. 2015. 〈hal-01198424〉

Partager

Métriques

Consultations de la notice

134