Neural Network Front-ends Based Speech Recognition in Reverberant Environments

DOI

10.2991/cset-16.2016.50How to use a DOI?

Keywords

REVERB Challenge, deep neural networks, Der-everberation, speech recognition

Abstract

This paper presents an investigation of reverberant speech recognition using frond-ends based methods. A 2-channel dereverberation method is adopted to achieve robust derever-beration under different reverberant conditions. Also a 2-channel spectral enhancement method is used where the gain of each frequency bin is controlled by acoustic scene, which is detected based on the analysis of full-band coherent property. Deep Neural Network (DNN) is also presented as a feature extractor. The DNN based front-end allows a very flexible integration of meta-information. Bottle neck features is extracted in place of MFCC features used in HMM-GMM system. We evaluated our methods on the data provided by REVERB challenge. On simulated data, the DNN front-end yields more than 33% relative reduction in Word Error Rate (WER).

Copyright

© 2016, the Authors. Published by Atlantis Press.

Open Access

This is an open access article distributed under the CC BY-NC license (http://creativecommons.org/licenses/by-nc/4.0/).

Download article (PDF)

TY  - CONF
AU  - Zhen Zhang
AU  - Peng Li
PY  - 2016/08
DA  - 2016/08
TI  - Neural Network Front-ends Based Speech Recognition in Reverberant Environments
BT  - Proceedings of the 2016 International Conference on Computer Science and Electronic Technology
PB  - Atlantis Press
SP  - 213
EP  - 218
SN  - 2352-538X
UR  - https://doi.org/10.2991/cset-16.2016.50
DO  - 10.2991/cset-16.2016.50
ID  - Zhang2016/08
ER  -