Publication Details

End-to-end DNN based text-independent speaker recognition for long and short utterances

ROHDIN, J.; SILNOVA, A.; DIEZ SÁNCHEZ, M.; PLCHOT, O.; MATĚJKA, P.; BURGET, L.; GLEMBEK, O. End-to-end DNN based text-independent speaker recognition for long and short utterances. COMPUTER SPEECH AND LANGUAGE, 2020, vol. 2020, no. 59, p. 22-35. ISSN: 0885-2308.
Czech title
Rozpoznávání mluvčího závislé na textu založené na End-to-end DNN přístupu pro dlouhé a krátké promluvy
Type
journal article
Language
English
Authors
Rohdin Johan Andréas, M.Sc., Ph.D. (DCGM)
Silnova Anna, M.Sc., Ph.D. (DCGM)
Diez Sánchez Mireia, M.Sc., Ph.D. (DCGM)
Plchot Oldřich, Ing., Ph.D. (DCGM)
Matějka Pavel, Ing., Ph.D. (DCGM)
Burget Lukáš, doc. Ing., Ph.D. (DCGM)
Glembek Ondřej, Ing., Ph.D.
URL
Keywords

Speaker verification, DNN, End-to-end, Text-independent, i-vector, PLDA

Abstract

Recently several end-to-end speaker verification systems based on deep neural networks (DNNs) have been proposed. These systems have been proven to be competitive for text-dependent tasks as well as for text-independent tasks with short utterances. However, for text-independent tasks with longer utterances, end-to-end systems are still outperformed by standard i-vector + PLDA systems. In this work, we present an end-to-end speaker verification system that is initialized to mimic an i-vector + PLDA baseline. The system is then further trained in an end-to-end manner but regularized so that it does not deviate too far from the initial system. In this way we mitigate overfitting which normally limits the performance of end-to-end systems. The proposed system outperforms the i-vector + PLDA baseline on both long and short duration utterances.

Published
2020
Pages
22–35
Journal
COMPUTER SPEECH AND LANGUAGE, vol. 2020, no. 59, ISSN 0885-2308
DOI
10.1016/j.csl.2019.06.002
UT WoS
000490540900002
EID Scopus
2-s2.0-85067618095
BibTeX 
@article{BUT158088,
  author="Johan Andréas {Rohdin} and Anna {Silnova} and Mireia {Diez Sánchez} and Oldřich {Plchot} and Pavel {Matějka} and Lukáš {Burget} and Ondřej {Glembek}",
  title="End-to-end DNN based text-independent speaker recognition for long and short utterances",
  journal="COMPUTER SPEECH AND LANGUAGE",
  year="2020",
  volume="2020",
  number="59",
  pages="22--35",
  doi="10.1016/j.csl.2019.06.002",
  issn="0885-2308",
  url="https://www.sciencedirect.com/science/article/pii/S0885230818303632"
}
Back to top