Exploring some limits of Gaussian PLDA modeling for i-vector distributions - Archive ouverte HAL Accéder directement au contenu
Communication Dans Un Congrès Année : 2014

Exploring some limits of Gaussian PLDA modeling for i-vector distributions

Résumé

Gaussian-PLDA (G-PLDA) modeling for i-vector based speaker verification has proven to be competitive versus heavy-tailed PLDA (HT-PLDA) based on Student's t-distribution, when the latter is much more computationally expensive. However , its results are achieved using a length-normalization, which projects i-vectors on the non-linear and finite surface of a hypersphere. This paper investigates the limits of linear and Gaussian G-PLDA modeling when distribution of data is spherical. In particular, assumptions of homoscedasticity are questionable: the model assumes that the within-speaker variability can be estimated by a unique and linear parameter. A non-probabilistic approach is proposed, competitive with state-of-the-art, which reveals some limits of the Gaussian modeling in terms of goodness of fit. We carry out an analysis of residue, which finds out a relation between the dispersion of a speaker-class and its location and, thus, shows that homoscedasticity assumptions are not fulfilled.

Mots clés

Fichier principal
Vignette du fichier
odyssey2014_submission_3.pdf (414.64 Ko) Télécharger le fichier
Origine : Fichiers produits par l'(les) auteur(s)
Loading...

Dates et versions

hal-02159801 , version 1 (20-06-2019)

Identifiants

  • HAL Id : hal-02159801 , version 1

Citer

Pierre-Michel Bousquet Bousquet, Jean-François Bonastre, Driss Matrouf. Exploring some limits of Gaussian PLDA modeling for i-vector distributions. Odyssey: The Speaker and Language Recognition Workshop, 2014, Joensuu, Finland. ⟨hal-02159801⟩

Collections

UNIV-AVIGNON LIA
45 Consultations
31 Téléchargements

Partager

Gmail Facebook X LinkedIn More