Constrained discriminative speaker verification specific to normalized i-vectors

Abstract : This paper focuses on discriminative trainings (DT) applied to i-vectors after Gaussian probabilistic linear discriminant analysis (PLDA). If DT has been successfully used with non-normalized vectors, this technique struggles to improve speaker detection when i-vectors have been first normalized, whereas the latter option has proven to achieve best performance in speaker verification. We propose an additional normalization procedure which limits the amount of coefficient to discriminatively train, with a minimal loss of accuracy. Adaptations of logistic regression based-DT to this new configuration are proposed, then we introduce a discriminative classifier for speaker verification which is a novelty in the field.
Document type :
Conference papers
Complete list of metadatas

Cited literature [26 references]  Display  Hide  Download

https://hal.archives-ouvertes.fr/hal-02159804
Contributor : Pierre-Michel Bousquet <>
Submitted on : Wednesday, June 19, 2019 - 9:06:30 AM
Last modification on : Friday, June 21, 2019 - 1:44:43 AM

File

Odyssey_2016_paper_3.pdf
Files produced by the author(s)

Identifiers

  • HAL Id : hal-02159804, version 1

Collections

Citation

Pierre-Michel Bousquet, Jean-François Bonastre. Constrained discriminative speaker verification specific to normalized i-vectors. Odyssey: The Speaker and Language Recognition Workshop, 2016, Bilbao, Spain. ⟨hal-02159804⟩

Share

Metrics

Record views

7

Files downloads

14