Publication detail

Full-covariance UBM and Heavy-tailed PLDA in I-Vector Speaker Verification

MATĚJKA, P. GLEMBEK, O. CASTALDO, F. ALAM, J. PLCHOT, O. KENNY, P. BURGET, L. ČERNOCKÝ, J.

Original Title

Full-covariance UBM and Heavy-tailed PLDA in I-Vector Speaker Verification

English Title

Full-covariance UBM and Heavy-tailed PLDA in I-Vector Speaker Verification

Type

conference paper

Language

en

Original Abstract

The work we presented aims at the best performance of the single stand alone system. We have presented full-covariance UBM and i-vector extraction with different kind of modeling. Our analysis shows that for the best performance it is necessary to have fullcovariance i-vector without any approximation.

English abstract

The work we presented aims at the best performance of the single stand alone system. We have presented full-covariance UBM and i-vector extraction with different kind of modeling. Our analysis shows that for the best performance it is necessary to have fullcovariance i-vector without any approximation.

Keywords

GMM, speaker recognition, PLDA, heavytailed PLDA, full-covariance UBM, i-vectors

RIV year

2011

Released

22.05.2011

Publisher

IEEE Signal Processing Society

Location

Praha

ISBN

978-1-4577-0537-3

Book

Proceedings of the 2011 IEEE International Conference on Acoustics, Speech, and Signal Processing, ICASSP 2011

Edition

NEUVEDEN

Edition number

NEUVEDEN

Pages from

4828

Pages to

4831

Pages count

4

URL

Documents

BibTex


@inproceedings{BUT76387,
  author="Pavel {Matějka} and Ondřej {Glembek} and Fabio {Castaldo} and Jahangir {Alam} and Oldřich {Plchot} and Patrick {Kenny} and Lukáš {Burget} and Jan {Černocký}",
  title="Full-covariance UBM and Heavy-tailed PLDA in I-Vector Speaker Verification",
  annote="The work we presented aims at the best performance of the single stand alone
system. We have presented full-covariance UBM and i-vector extraction with
different kind of modeling. Our analysis shows that for the best performance it
is necessary to have fullcovariance i-vector without any approximation.",
  address="IEEE Signal Processing Society",
  booktitle="Proceedings of the 2011 IEEE International Conference on Acoustics, Speech, and Signal Processing, ICASSP 2011",
  chapter="76387",
  doi="10.1109/ICASSP.2011.5947436",
  edition="NEUVEDEN",
  howpublished="print",
  institution="IEEE Signal Processing Society",
  year="2011",
  month="may",
  pages="4828--4831",
  publisher="IEEE Signal Processing Society",
  type="conference paper"
}