Publication detail

Discriminatively Trained i-vector Extractor for Speaker Verification

GLEMBEK, O. BURGET, L. BRÜMMER, N. PLCHOT, O. MATĚJKA, P.

Original Title

Discriminatively Trained i-vector Extractor for Speaker Verification

Type

conference paper

Language

English

Original Abstract

We have proposed a technique for discriminative training of the i-vector extractor parameters using cross-entropy as the error function. We have applied the technique both to the original i-vector extractor and to its simplified version. In both cases, the discriminative training was effective, giving higher relative improvement in the simplified case.

Keywords

speaker verification, i-vectors, PLDA, discriminative training

Authors

GLEMBEK, O.; BURGET, L.; BRÜMMER, N.; PLCHOT, O.; MATĚJKA, P.

RIV year

2011

Released

27. 8. 2011

Publisher

International Speech Communication Association

Location

Florence

ISBN

978-1-61839-270-1

Book

Proceedings of Interspeech 2011

ISBN

1990-9772

Periodical

Proceedings of Interspeech

Year of study

2011

Number

8

State

French Republic

Pages from

137

Pages to

140

Pages count

4

URL

BibTex

@inproceedings{BUT76447,
  author="Ondřej {Glembek} and Lukáš {Burget} and Niko {Brümmer} and Oldřich {Plchot} and Pavel {Matějka}",
  title="Discriminatively Trained i-vector Extractor for Speaker Verification",
  booktitle="Proceedings of Interspeech 2011",
  year="2011",
  journal="Proceedings of Interspeech",
  volume="2011",
  number="8",
  pages="137--140",
  publisher="International Speech Communication Association",
  address="Florence",
  isbn="978-1-61839-270-1",
  issn="1990-9772",
  url="http://www.fit.vutbr.cz/research/groups/speech/publi/2011/glembek_interspeech2011_386.pdf"
}