DMCA
Probabilistic Linear Discriminant Analysis with Bottleneck Features for Speech Recognition
| Citations: | 6 - 3 self |
BibTeX
@MISC{Lu_probabilisticlinear,
author = {Liang Lu and Steve Renals},
title = {Probabilistic Linear Discriminant Analysis with Bottleneck Features for Speech Recognition},
year = {}
}
Share
 |
 |
 |
 |
||
OpenURL
Abstract
We have recently proposed a new acoustic model based on prob-abilistic linear discriminant analysis (PLDA) which enjoys the flexibility of using higher dimensional acoustic features, and is more capable to capture the intra-frame feature correlations. In this paper, we investigate the use of bottleneck features obtained from a deep neural network (DNN) for the PLDA-based acous-tic model. Experiments were performed on the Switchboard dataset — a large vocabulary conversational telephone speech corpus. We observe significant word error reduction by using the bottleneck features. In addition, we have also compared the PLDA-based acoustic model to three others using Gaussian mixture models (GMMs), subspace GMMs and hybrid deep neural networks (DNNs), and PLDA can achieve comparable or slightly higher recognition accuracy from our experiments. Index Terms: speech recognition, bottleneck features, proba-bilistic linear discriminant analysis
Keyphrases
bottleneck feature speech recognition probabilistic linear discriminant analysis index term proba-bilistic linear discriminant analysis gaussian mixture model plda-based acoustic model intra-frame feature correlation plda-based acous-tic model recognition accuracy hybrid deep neural network prob-abilistic linear discriminant analysis new acoustic model significant word error reduction subspace gmms deep neural network dimensional acoustic feature
