Loading...
Please wait, while we are loading the content...
Similar Documents
Feature Learning with Gaussian Restricted Boltzmann Machine for Robust Speech Recognition
| Content Provider | CiteSeerX |
|---|---|
| Author | Zheng, Xin Wu, Zhiyong Meng, Helen Li, Weifeng Cai, Lianhong |
| Abstract | In this paper, we first present a new variant of Gaussian re-stricted Boltzmann machine (GRBM) called multivariate Gaus-sian restricted Boltzmann machine (MGRBM), with its def-inition and learning algorithm. Then we propose using a learned GRBM or MGRBM to extract better features for ro-bust speech recognition. Our experiments on Aurora2 show that both GRBM-extracted and MGRBM-extracted feature per-forms much better than Mel-frequency cepstral coefficient (MFCC) with either HMM-GMM or hybrid HMM-deep neural network (DNN) acoustic model, and MGRBM-extracted fea-ture is slightly better. Index Terms: restricted Boltzmann machine, robust speech recognition, feature learning |
| File Format | |
| Access Restriction | Open |
| Content Type | Text |
