We are sorry, an unexpected error has occured. Please, contact portal@vutbr.cz and copy this error:
Error in SQL statement:
Tabulka nebo pohled neexistuje [942]

Detail publikačního výsledku

Factorization of Discriminatively Trained i-Vector Extractor for Speaker Recognition

NOVOTNÝ, O.; PLCHOT, O.; GLEMBEK, O.; BURGET, L.

Original Title

Factorization of Discriminatively Trained i-Vector Extractor for Speaker Recognition

English Title

Factorization of Discriminatively Trained i-Vector Extractor for Speaker Recognition

Type

Paper in proceedings (conference paper)

Original Abstract

In this work, we continue in our research on i-vector extractorfor speaker verification (SV) and we optimize its architecturefor fast and effective discriminative training. We were motivatedby computational and memory requirements caused bythe large number of parameters of the original generative ivectormodel. Our aim is to preserve the power of the originalgenerative model, and at the same time focus the model towardsextraction of speaker-related information. We show that it ispossible to represent a standard generative i-vector extractor bya model with significantly less parameters and obtain similarperformance on SV tasks. We can further refine this compactmodel by discriminative training and obtain i-vectors that leadto better performance on various SV benchmarks representingdifferent acoustic domains.

English abstract

In this work, we continue in our research on i-vector extractorfor speaker verification (SV) and we optimize its architecturefor fast and effective discriminative training. We were motivatedby computational and memory requirements caused bythe large number of parameters of the original generative ivectormodel. Our aim is to preserve the power of the originalgenerative model, and at the same time focus the model towardsextraction of speaker-related information. We show that it ispossible to represent a standard generative i-vector extractor bya model with significantly less parameters and obtain similarperformance on SV tasks. We can further refine this compactmodel by discriminative training and obtain i-vectors that leadto better performance on various SV benchmarks representingdifferent acoustic domains.

Keywords

SRE

Key words in English

SRE

Authors

NOVOTNÝ, O.; PLCHOT, O.; GLEMBEK, O.; BURGET, L.

RIV year

2020

Released

15.09.2019

Publisher

International Speech Communication Association

Location

Graz

Book

Proceedings of Interspeech

ISBN

1990-9772

Periodical

Proceedings of Interspeech

Volume

2019

Number

9

State

French Republic

Pages from

4330

Pages to

4334

Pages count

5

URL

https://www.isca-speech.org/archive/Interspeech_2019/pdfs/1757.pdf

Full text in the Digital Library

http://hdl.handle.net/

BibTex

@inproceedings{BUT159998,
  author="Ondřej {Novotný} and Oldřich {Plchot} and Ondřej {Glembek} and Lukáš {Burget}",
  title="Factorization of Discriminatively Trained i-Vector Extractor for Speaker Recognition",
  booktitle="Proceedings of Interspeech",
  year="2019",
  journal="Proceedings of Interspeech",
  volume="2019",
  number="9",
  pages="4330--4334",
  publisher="International Speech Communication Association",
  address="Graz",
  doi="10.21437/Interspeech.2019-1757",
  issn="1990-9772",
  url="https://www.isca-speech.org/archive/Interspeech_2019/pdfs/1757.pdf"
}

Documents

novotny_is2019_191757