Result Details

End-to-End DNN Based Speaker Recognition Inspired by i-Vector and PLDA

ROHDIN, J.; SILNOVA, A.; DIEZ SÁNCHEZ, M.; PLCHOT, O.; MATĚJKA, P.; BURGET, L. End-to-End DNN Based Speaker Recognition Inspired by i-Vector and PLDA. In Proceedings of ICASSP. Calgary: IEEE Signal Processing Society, 2018. p. 4874-4878. ISBN: 978-1-5386-4658-8.
Type
conference paper
Language
English
Authors
Rohdin Johan Andréas, M.Sc., Ph.D., FIT (FIT), DCGM (FIT)
Silnova Anna, M.Sc., Ph.D., DCGM (FIT)
Diez Sánchez Mireia, M.Sc., Ph.D., DCGM (FIT)
Plchot Oldřich, Ing., Ph.D., DCGM (FIT)
Matějka Pavel, Ing., Ph.D., DCGM (FIT)
Burget Lukáš, doc. Ing., Ph.D., DCGM (FIT)
Abstract

Recently, several end-to-end speaker verification systems based ondeep neural networks (DNNs) have been proposed. These systemshave been proven to be competitive for text-dependent tasks as wellas for text-independent tasks with short utterances. However, fortext-independent tasks with longer utterances, end-to-end systemsare still outperformed by standard i-vector + PLDA systems. In thiswork, we develop an end-to-end speaker verification system that isinitialized to mimic an i-vector + PLDA baseline. The system isthen further trained in an end-to-end manner but regularized so thatit does not deviate too far from the initial system. In this way wemitigate overfitting which normally limits the performance of endto-end systems. The proposed system outperforms the i-vector +PLDA baseline on both long and short duration utterances.

Keywords

Speaker verification, DNN, end-to-end

URL
Published
2018
Pages
4874–4878
Proceedings
Proceedings of ICASSP
Conference
IEEE International Conference on Acoustics, Speech and Signal Processing
ISBN
978-1-5386-4658-8
Publisher
IEEE Signal Processing Society
Place
Calgary
DOI
10.1109/ICASSP.2018.8461958
UT WoS
000446384605009
EID Scopus
2-s2.0-85054212885
BibTeX 
@inproceedings{BUT155046,
  author="Johan Andréas {Rohdin} and Anna {Silnova} and Mireia {Diez Sánchez} and Oldřich {Plchot} and Pavel {Matějka} and Lukáš {Burget}",
  title="End-to-End DNN Based Speaker Recognition Inspired by i-Vector and PLDA",
  booktitle="Proceedings of ICASSP",
  year="2018",
  pages="4874--4878",
  publisher="IEEE Signal Processing Society",
  address="Calgary",
  doi="10.1109/ICASSP.2018.8461958",
  isbn="978-1-5386-4658-8",
  url="https://www.fit.vut.cz/research/publication/11724/"
}
Files
Projects
Improving Robustnes in Automatic Speaker Recognition, GACR, Juniorské granty, GJ17-23870Y, start: 2017-01-01, end: 2019-12-31, completed
IT4Innovations excellence in science, MŠMT, Národní program udržitelnosti II, LQ1602, start: 2016-01-01, end: 2020-12-31, completed
Neural networks for signal processing and speech data mining, TAČR, Program na podporu aplikovaného výzkumu ZÉTA, TJ01000208, start: 2018-01-01, end: 2019-12-31, completed
NTT - Speech enhancement front-end for robust automatic speech recognition with large amount of training data, NTT, start: 2017-10-01, end: 2018-09-30, completed
Robust SPEAKER DIariazation systems using Bayesian inferenCE and deep learning methods, EU, Horizon 2020, start: 2017-03-01, end: 2019-02-28, completed
Sequence summarizing neural networks for speaker recognition, EU, Horizon 2020, 5SA15094, start: 2016-07-01, end: 2019-06-30, completed
Research groups
Speech Data Mining Research Group BUT Speech@FIT (RG SPEECH)
Departments
Department of Computer Graphics and Multimedia (DCGM)
Back to top