Detail publikace

But System for the Second Dihard Speech Diarization Challenge

LANDINI, F. WANG, S. DIEZ SÁNCHEZ, M. BURGET, L. MATĚJKA, P. ŽMOLÍKOVÁ, K. MOŠNER, L. SILNOVA, A. PLCHOT, O. NOVOTNÝ, O. ZEINALI, H. ROHDIN, J.

Originální název

But System for the Second Dihard Speech Diarization Challenge

Anglický název

But System for the Second Dihard Speech Diarization Challenge

Jazyk

en

Originální abstrakt

This paper describes the winning systems developed by the BUT team for the four tracks of the Second DIHARD Speech Diarization Challenge. For tracks 1 and 2 the systems were mainly based on performing agglomerative hierarchical clustering (AHC) of x-vectors, followed by another x-vector clustering based on Bayes hidden Markov model and variational Bayes inference. We provide a comparison of the improvement given by each step and share the implementation of the core of the system. For tracks 3 and 4 with recordings from the Fifth CHiME Challenge, we explored different approaches for doing multi-channel diarization and our best performance was obtained when applying AHC on the fusion of per channel probabilistic linear discriminant analysis scores.

Anglický abstrakt

This paper describes the winning systems developed by the BUT team for the four tracks of the Second DIHARD Speech Diarization Challenge. For tracks 1 and 2 the systems were mainly based on performing agglomerative hierarchical clustering (AHC) of x-vectors, followed by another x-vector clustering based on Bayes hidden Markov model and variational Bayes inference. We provide a comparison of the improvement given by each step and share the implementation of the core of the system. For tracks 3 and 4 with recordings from the Fifth CHiME Challenge, we explored different approaches for doing multi-channel diarization and our best performance was obtained when applying AHC on the fusion of per channel probabilistic linear discriminant analysis scores.

Dokumenty

BibTex


@inproceedings{BUT163962,
  author="Federico Nicolás {Landini} and Shuai {Wang} and Mireia {Diez Sánchez} and Lukáš {Burget} and Pavel {Matějka} and Kateřina {Žmolíková} and Ladislav {Mošner} and Anna {Silnova} and Oldřich {Plchot} and Ondřej {Novotný} and Hossein {Zeinali} and Johan Andréas {Rohdin}",
  title="But System for the Second Dihard Speech Diarization Challenge",
  annote="This paper describes the winning systems developed by the BUT team for the four
tracks of the Second DIHARD Speech Diarization Challenge. For tracks 1 and 2 the
systems were mainly based on performing agglomerative hierarchical clustering
(AHC) of x-vectors, followed by another x-vector clustering based on Bayes hidden
Markov model and variational Bayes inference. We provide a comparison of the
improvement given by each step and share the implementation of the core of the
system. For tracks 3 and 4 with recordings from the Fifth CHiME Challenge, we
explored different approaches for doing multi-channel diarization and our best
performance was obtained when applying AHC on the fusion of per channel
probabilistic linear discriminant analysis scores.",
  address="IEEE Signal Processing Society",
  booktitle="ICASSP, IEEE International Conference on Acoustics, Speech and Signal Processing - Proceedings",
  chapter="163962",
  doi="10.1109/ICASSP40776.2020.9054251",
  edition="NEUVEDEN",
  howpublished="online",
  institution="IEEE Signal Processing Society",
  year="2020",
  month="may",
  pages="6529--6533",
  publisher="IEEE Signal Processing Society",
  type="conference paper"
}