PLDA modeling in the fishervoice subspace for speaker verification

Jinghua Zhong, Weiwu Jiang, Wei Rao, Man Wai Mak, Helen Meng

Research output: Journal article publicationConference articleAcademic researchpeer-review

1 Citation (Scopus)

Abstract

We have previously developed a Fishervoice framework that maps the JFA-mean supervectors into a compressed discriminant subspace using nonparametric Fishers discriminant analysis. It was shown that performing cosine distance scoring (CDS) on these Fishervoice projected vectors (denoted as f-vectors) can outperform the classical joint factor analysis. Unlike the ivector approach in which the channel variability is suppressed in the classification stage, in the Fishervoice framework, channel variability is suppressed when the f-vectors are constructed. In this paper, we investigate whether channel variability can be further suppressed by performing Gaussian probabilistic discriminant analysis (PLDA) in the classification stage. We also use random subspace sampling to enrich the speaker discriminative information in the f-vectors. Experiments on NIST SRE10 show that PLDA can boost the performance of Fishervoice in speaker verification significantly by a relative decrease of 14.4% in minDCF (from 0.526 to 0.450).
Original languageEnglish
Pages (from-to)1130-1134
Number of pages5
JournalProceedings of the Annual Conference of the International Speech Communication Association, INTERSPEECH
Publication statusPublished - 1 Jan 2014
Event15th Annual Conference of the International Speech Communication Association: Celebrating the Diversity of Spoken Languages, INTERSPEECH 2014 - Max Atria at Singapore Expo, Singapore, Singapore
Duration: 14 Sep 201418 Sep 2014

Keywords

  • Fishervoice
  • Joint factor analysis
  • Probabilistic linear discriminant analysis
  • Random sampling
  • Supervector

ASJC Scopus subject areas

  • Language and Linguistics
  • Human-Computer Interaction
  • Signal Processing
  • Software
  • Modelling and Simulation

Cite this