Publication Details

Fast variational Bayes for heavy-tailed PLDA applied to i-vectors and x-vectors

SILNOVA Anna, BRUMMER Niko, GARCÍA-ROMERO Daniel, SNYDER David and BURGET Lukáš. Fast variational Bayes for heavy-tailed PLDA applied to i-vectors and x-vectors. In: Proceedings of Interspeech 2018. Hyderabad: International Speech Communication Association, 2018, pp. 72-76. ISSN 1990-9772. Available from: https://www.isca-speech.org/archive/Interspeech_2018/abstracts/2128.html
Czech title
Rychlý variační Bayes pro PLDA model s těžkým chvostem aplikovaný na i-vektory a x-vektory
Type
conference paper
Language
english
Authors
Silnova Anna, MSc. (DCGM FIT BUT)
Brummer Niko (OMILIA)
García-Romero Daniel (JHU)
Snyder David (JHU)
Burget Lukáš, doc. Ing., Ph.D. (DCGM FIT BUT)
URL
Keywords

peaker recognition, variational Bayes, heavytailed PLDA

Abstract

The standard state-of-the-art backend for text-independent speaker recognizers that use i-vectors or x-vectors, is Gaussian PLDA (G-PLDA), assisted by a Gaussianization step involving length normalization. G-PLDA can be trained with both generative or discriminative methods. It has long been known that heavy-tailed PLDA (HT-PLDA), applied without length normalization, gives similar accuracy, but at considerable extra computational cost. We have recently introduced a fast scoring algorithm for a discriminatively trained HT-PLDA backend. This paper extends that work by introducing a fast, variational Bayes, generative training algorithm. We compare old and new backends, with and without length-normalization, with i-vectors and x-vectors, on SRE10, SRE16 and SITW.

Published
2018
Pages
72-76
Journal
Proceedings of Interspeech, vol. 2018, no. 9, ISSN 1990-9772
Proceedings
Proceedings of Interspeech 2018
Conference
19TH ANNUAL CONFERENCE OF THE INTERNATIONAL SPEECH COMMUNICATION ASSOCIATION (INTERSPEECH 2018), Hyderabad, India, IN
Publisher
International Speech Communication Association
Place
Hyderabad, IN
DOI
UT WoS
000465363900015
EID Scopus
BibTeX
@INPROCEEDINGS{FITPUB11837,
   author = "Anna Silnova and Niko Brummer and Daniel Garc\'{i}a-Romero and David Snyder and Luk\'{a}\v{s} Burget",
   title = "Fast variational Bayes for heavy-tailed PLDA applied to i-vectors and x-vectors",
   pages = "72--76",
   booktitle = "Proceedings of Interspeech 2018",
   journal = "Proceedings of Interspeech",
   volume = 2018,
   number = 9,
   year = 2018,
   location = "Hyderabad, IN",
   publisher = "International Speech Communication Association",
   ISSN = "1990-9772",
   doi = "10.21437/Interspeech.2018-2128",
   language = "english",
   url = "https://www.fit.vut.cz/research/publication/11837"
}
Back to top