ISCA Archive Interspeech 2013
ISCA Archive Interspeech 2013

Mixtures of Bayesian joint factor analyzers for noise robust automatic speech recognition

Xiaodong Cui, Vaibhava Goel, Brian Kingsbury

This paper investigates a noise robust approach to automatic speech recognition based on a mixture of Bayesian joint factor analyzers. In this approach, noisy features are modeled by two joint groups of factors accounting for speaker and noise variabilities which are estimated by clean and noisy speech respectively. The factors form an overcomplete dictionary with a redundant representation. Automatic relevance determination (ARD) is carried out by the relevance vector machine (RVM) where sparsity-promoting priors are applied on two factor loading matrices. Experiments on large vocabulary continuous speech recognition (LVCSR) tasks show good improvements by this approach.


doi: 10.21437/Interspeech.2013-280

Cite as: Cui, X., Goel, V., Kingsbury, B. (2013) Mixtures of Bayesian joint factor analyzers for noise robust automatic speech recognition. Proc. Interspeech 2013, 3012-3016, doi: 10.21437/Interspeech.2013-280

@inproceedings{cui13_interspeech,
  author={Xiaodong Cui and Vaibhava Goel and Brian Kingsbury},
  title={{Mixtures of Bayesian joint factor analyzers for noise robust automatic speech recognition}},
  year=2013,
  booktitle={Proc. Interspeech 2013},
  pages={3012--3016},
  doi={10.21437/Interspeech.2013-280}
}