Detail publikace

Analyzing speaker verification embedding extractors and back-ends under language and channel mismatch

SILNOVA, A. STAFYLAKIS, T. MOŠNER, L. PLCHOT, O. ROHDIN, J. MATĚJKA, P. BURGET, L. GLEMBEK, O. BRUMMER, J.

Originální název

Analyzing speaker verification embedding extractors and back-ends under language and channel mismatch

Typ

článek ve sborníku mimo WoS a Scopus

Jazyk

angličtina

Originální abstrakt

In this paper, we analyze the behavior and performance of speaker embeddings and the back-end scoring model under domain and language mismatch. We present our findings regarding ResNet-based speaker embedding architectures and show that reduced temporal stride yields improved performance. We then consider a PLDA back-end and show how a combination of small speaker subspace, language-dependent PLDA mixture, and nuisance-attribute projection can have a drastic impact on the performance of the system. Besides, we present an efficient way of scoring and fusing class posterior logit vectors recently shown to perform well on speaker verification task. The experiments are performed using the NIST SRE 2021 setup.

Klíčová slova

speaker, verification, embedding

Autoři

SILNOVA, A.; STAFYLAKIS, T.; MOŠNER, L.; PLCHOT, O.; ROHDIN, J.; MATĚJKA, P.; BURGET, L.; GLEMBEK, O.; BRUMMER, J.

Vydáno

28. 6. 2022

Nakladatel

International Speech Communication Association

Místo

Beijing

Strany od

9

Strany do

16

Strany počet

8

URL

https://www.isca-speech.org/archive/pdfs/odyssey_2022/silnova22_odyssey.pdf

BibTex

@inproceedings{BUT179660,
  author="Anna {Silnova} and Themos {Stafylakis} and Ladislav {Mošner} and Oldřich {Plchot} and Johan Andréas {Rohdin} and Pavel {Matějka} and Lukáš {Burget} and Ondřej {Glembek} and Johan Nikolaas Langenhoven {Brummer}",
  title="Analyzing speaker verification embedding extractors and back-ends under language and channel mismatch",
  booktitle="Proceedings of The Speaker and Language Recognition Workshop (Odyssey 2022)",
  year="2022",
  pages="9--16",
  publisher="International Speech Communication Association",
  address="Beijing",
  doi="10.21437/Odyssey.2022-2",
  url="https://www.isca-speech.org/archive/pdfs/odyssey_2022/silnova22_odyssey.pdf"
}

Dokumenty

silnova_odyssey2022_analysis.pdf