Detail publikačního výsledku

Speaker Diarization based on Bayesian HMM with Eigenvoice Priors

DIEZ SÁNCHEZ, M.; BURGET, L.; MATĚJKA, P.

Originální název

Speaker Diarization based on Bayesian HMM with Eigenvoice Priors

Anglický název

Speaker Diarization based on Bayesian HMM with Eigenvoice Priors

Druh

Stať ve sborníku v databázi WoS či Scopus

Originální abstrakt

Nowadays, most speaker diarization methods address thetask in two steps: segmentation of the input conversation into(preferably) speaker homogeneous segments, and clustering.Generally, different models and techniques are used for the twosteps. In this paper we present a very elegant approach where astraightforward and efficient Variational Bayes (VB) inferencein a single probabilistic model addresses the complete SD problem.Our model is a Bayesian Hidden Markov Model, in whichstates represent speaker specific distributions and transitions betweenstates represent speaker turns. As in the ivector or JFAmodels, speaker distributions are modeled by GMMs with parametersconstrained by eigenvoice priors. This allows to robustlyestimate the speaker models from very short speech segments.The model, which was released as open source codeand has already been used by several labs, is fully describedfor the first time in this paper. We present results and the systemis compared and combined with other state-of-the-art approaches.The model provides the best results reported so faron the CALLHOME dataset.

Anglický abstrakt

Nowadays, most speaker diarization methods address thetask in two steps: segmentation of the input conversation into(preferably) speaker homogeneous segments, and clustering.Generally, different models and techniques are used for the twosteps. In this paper we present a very elegant approach where astraightforward and efficient Variational Bayes (VB) inferencein a single probabilistic model addresses the complete SD problem.Our model is a Bayesian Hidden Markov Model, in whichstates represent speaker specific distributions and transitions betweenstates represent speaker turns. As in the ivector or JFAmodels, speaker distributions are modeled by GMMs with parametersconstrained by eigenvoice priors. This allows to robustlyestimate the speaker models from very short speech segments.The model, which was released as open source codeand has already been used by several labs, is fully describedfor the first time in this paper. We present results and the systemis compared and combined with other state-of-the-art approaches.The model provides the best results reported so faron the CALLHOME dataset.

Klíčová slova

Speaker diarization, speaker recognition

Klíčová slova v angličtině

Speaker diarization, speaker recognition

Autoři

DIEZ SÁNCHEZ, M.; BURGET, L.; MATĚJKA, P.

Rok RIV

2019

Vydáno

26.06.2018

Nakladatel

International Speech Communication Association

Místo

Les Sables d´Olonne

Kniha

Proceedings of Odyssey 2018

ISSN

2312-2846

Periodikum

Proceedings of Odyssey: The Speaker and Language Recognition Workshop Odyssey 2014, Joensuu, Finland

Svazek

2018

Číslo

6

Stát

Finská republika

Strany od

147

Strany do

154

Strany počet

8

URL

https://www.fit.vut.cz/research/publication/11786/

Plný text v Digitální knihovně

http://hdl.handle.net/

BibTex

@inproceedings{BUT155067,
  author="Mireia {Diez Sánchez} and Lukáš {Burget} and Pavel {Matějka}",
  title="Speaker Diarization based on Bayesian HMM with Eigenvoice Priors",
  booktitle="Proceedings of Odyssey 2018",
  year="2018",
  journal="Proceedings of Odyssey: The Speaker and Language Recognition Workshop Odyssey 2014, Joensuu, Finland",
  volume="2018",
  number="6",
  pages="147--154",
  publisher="International Speech Communication Association",
  address="Les Sables d´Olonne",
  doi="10.21437/Odyssey.2018-21",
  issn="2312-2846",
  url="https://www.fit.vut.cz/research/publication/11786/"
}

Dokumenty

diez_odyssey2018_63