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

Abstract : 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.
Keywords : Speaker recognition
Document type :
Conference papers
Complete list of metadatas

Cited literature [19 references]  Display  Hide  Download

https://hal.archives-ouvertes.fr/hal-02159801
Contributor : Pierre-Michel Bousquet <>
Submitted on : Thursday, June 20, 2019 - 5:28:57 PM
Last modification on : Tuesday, July 2, 2019 - 5:38:02 PM

File

odyssey2014_submission_3.pdf
Files produced by the author(s)

Identifiers

  • HAL Id : hal-02159801, version 1

Collections

Citation

Pierre-Michel 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⟩

Share

Metrics

Record views

8

Files downloads

9