Publication Details

Integration of Variational Autoencoder and Spatial Clustering for Adaptive Multi-Channel Neural Speech Separation

ŽMOLÍKOVÁ, K.; DELCROIX, M.; BURGET, L.; NAKATANI, T.; ČERNOCKÝ, J. Integration of Variational Autoencoder and Spatial Clustering for Adaptive Multi-Channel Neural Speech Separation. In 2021 IEEE Spoken Language Technology Workshop, SLT 2021 - Proceedings. Shenzhen - virtual: IEEE Signal Processing Society, 2021. p. 889-896. ISBN: 978-1-7281-7066-4.
Czech title
Integrace variačního autoenkodéru a prostorového shlukování pro adaptivní multikanálovou neurální separaci řeči
Type
conference paper
Language
English
Authors
URL
Keywords

Multi-channel speech separation, variational autoencoder, spatial clustering,
DOLPHIN

Abstract

In this paper, we propose a method combining variational autoencoder model of
speech with a spatial clustering approach for multichannel speech separation. The
advantage of integrating spatial clustering with a spectral model was shown in
several works. As the spectral model, previous works used either factorial
generative models of the mixed speech or discriminative neural networks. In our
work, we combine the strengths of both approaches, by building a factorial model
based on a generative neural network, a variational autoencoder. By doing so, we
can exploit the modeling power of neural networks, but at the same time, keep
a structured model. Such a model can be advantageous when adapting to new noise
conditions as only the noise part of the model needs to be modified. We show
experimentally, that our model significantly outperforms previous factorial model
based on Gaussian mixture model (DOLPHIN), performs comparably to integration of
permutation invariant training with spatial clustering, and enables us to easily
adapt to new noise conditions.

Published
2021
Pages
889–896
Proceedings
2021 IEEE Spoken Language Technology Workshop, SLT 2021 - Proceedings
Conference
2021 IEEE Spoken Language Technology Workshop (SLT), Shenzhen - virtual conference, CN
ISBN
978-1-7281-7066-4
Publisher
IEEE Signal Processing Society
Place
Shenzhen - virtual
DOI
UT WoS
000663633300121
EID Scopus
BibTeX
@inproceedings{BUT175809,
  author="Kateřina {Žmolíková} and Marc {Delcroix} and Lukáš {Burget} and Tomohiro {Nakatani} and Jan {Černocký}",
  title="Integration of Variational Autoencoder and Spatial Clustering for Adaptive Multi-Channel Neural Speech Separation",
  booktitle="2021 IEEE Spoken Language Technology Workshop, SLT 2021 - Proceedings",
  year="2021",
  pages="889--896",
  publisher="IEEE Signal Processing Society",
  address="Shenzhen - virtual",
  doi="10.1109/SLT48900.2021.9383612",
  isbn="978-1-7281-7066-4",
  url="https://ieeexplore.ieee.org/document/9383612"
}
Files
Back to top