Sound event detection in remote health care - Small learning datasets and over constrained Gaussian Mixture Models - Archive ouverte HAL Accéder directement au contenu
Communication Dans Un Congrès Année : 2010

Sound event detection in remote health care - Small learning datasets and over constrained Gaussian Mixture Models

Résumé

The use of Gaussian Mixture Models (GMM), adapted through the Expectation Minimization(EM)algorithm, is not rare in Audio Analysis for Surveillance Applications and Environmental sound recognition. Their use, at a first glance, is founded on the good qualities of GMM models when aimed at approximating Probability Density Functions(PDF) of random variables. But in some cases, where models are to be adapted from small sample sets of specific and locally recorded signals, instead of large but generic databases, a problem of balance between model complexity and sample size may play an important role. From this perspective, we show, through simple sound classification experiments, that constrained GMM, with fewer degrees of freedom, as compared to GMM with full covariance matrices, provide better classification performances. Moreover, pushing this argument even further, we also show that a Parzen model (seen here as an over-constrained GMM) can do even better than usual GMM, in terms of classification error ratio
Fichier principal
Vignette du fichier
EMBS2010-Jugurta.pdf (134.4 Ko) Télécharger le fichier
Origine : Fichiers produits par l'(les) auteur(s)
Loading...

Dates et versions

hal-01308019 , version 1 (27-04-2016)

Identifiants

Citer

Jugurta Montalvão, Dan Istrate, Jérôme Boudy, Joan Mouba. Sound event detection in remote health care - Small learning datasets and over constrained Gaussian Mixture Models. EMBC 2010 : 32nd Annual International Conference of the IEEE Engineering in Medicine and Biology Society, Aug 2010, Buenos Aires, Argentina. pp.1146-1149, ⟨10.1109/IEMBS.2010.5627149⟩. ⟨hal-01308019⟩
51 Consultations
188 Téléchargements

Altmetric

Partager

Gmail Facebook X LinkedIn More