Publication detail

Analysis of Speaker Diarization based on Bayesian HMM with Eigenvoice Priors

DIEZ SÁNCHEZ, M. BURGET, L. LANDINI, F. ČERNOCKÝ, J.

Original Title

Analysis of Speaker Diarization based on Bayesian HMM with Eigenvoice Priors

Type

journal article in Web of Science

Language

English

Original Abstract

In our previous work, we introduced our Bayesian Hidden Markov Model with eigenvoice priors, which has been recently recognized as the state-of-the-art model for Speaker Diarization. In this paper we present a more complete analysis of the Diarization system. The inference of the model is fully described and derivations of all update formulas are provided for a complete understanding of the algorithm. An extensive analysis on the effect, sensitivity and interactions of all model parameters is provided, which might be used as a guide for their optimal setting. The newly introduced speaker regularization coefficient allows us to control the number of speakers inferred in an utterance. A naive speaker model merging strategy is also presented, which allows to drive the variational inference out of local optima. Experiments for the different diarization scenarios are presented on CALLHOME and DIHARD datasets.

Keywords

Hidden Markov Models, Bayes methods, Task analysis, Probabilistic logic, Training, Speech processing, Complexity theory

Authors

DIEZ SÁNCHEZ, M.; BURGET, L.; LANDINI, F.; ČERNOCKÝ, J.

Released

1. 12. 2020

ISBN

2329-9290

Periodical

IEEE/ACM TRANSACTIONS ON AUDIO, SPEECH AND LANGUAGE PROCESSING

Year of study

28

Number

1

State

United States of America

Pages from

355

Pages to

368

Pages count

14

URL

https://ieeexplore.ieee.org/document/8910412

BibTex

@article{BUT161472,
  author="Mireia {Diez Sánchez} and Lukáš {Burget} and Federico Nicolás {Landini} and Jan {Černocký}",
  title="Analysis of Speaker Diarization based on Bayesian HMM with Eigenvoice Priors",
  journal="IEEE/ACM TRANSACTIONS ON AUDIO, SPEECH AND LANGUAGE PROCESSING",
  year="2020",
  volume="28",
  number="1",
  pages="355--368",
  doi="10.1109/TASLP.2019.2955293",
  issn="2329-9290",
  url="https://ieeexplore.ieee.org/document/8910412"
}

Documents

MDiez_IEEE_TASLP_2020.pdf