The paper presents a probabilistic echo-state network (pi-ESN) for density estimation over variable-length sequences of multivariate random vectors. The pi-ESN sterns from the combination of the reservoir of an ESN and a parametric density model based on radial basis functions. A constrained maximum likelihood training algorithm is introduced, suitable for sequence classification. Extensions of the algorithm to unsupervised clustering and semi-supervised learning (SSL) of sequences are proposed. Experiments in emotion recognition from speech signals are conducted on the WaSeP (c) dataset. Compared with established techniques, the pi-ESN yields the highest recognition accuracies, and shows interesting clustering and SSL capabilities. (C) 2014 Elsevier B.V. All rights reserved.
Scheda prodotto non validato
Scheda prodotto in fase di analisi da parte dello staff di validazione
|Titolo:||Emotion recognition from speech signals via a probabilistic echo-state network|
|Citazione:||Trentin, E., Stefan, S., & Friedhelm, S. (2015). Emotion recognition from speech signals via a probabilistic echo-state network. PATTERN RECOGNITION LETTERS, 66, 4-12.|
|Appare nelle tipologie:||1.1 Articolo in rivista|