Multi-resolution spectral entropy feature for robust ASR

Recently, entropy measures at different stages of recognition have been used in automatic speech recognition (ASR) tasks. In a recent paper, we proposed that formant positions of a spectrum can be captured by a multi-resolution spectral entropy feature. In this paper, we suggest modifications to the...

Full description

Saved in:
Bibliographic Details
Main Authors: Misra, H., Ikbal, S., Sivadas, S., Bourlard, H.
Format: Conference Proceeding
Language:English
Subjects:
Acoustic noise
Additive noise
Automatic speech recognition
Cepstral analysis
Entropy
Hidden Markov models
Mel frequency cepstral coefficient
Multiresolution analysis
Noise robustness
Wideband
Online Access:Request full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:Recently, entropy measures at different stages of recognition have been used in automatic speech recognition (ASR) tasks. In a recent paper, we proposed that formant positions of a spectrum can be captured by a multi-resolution spectral entropy feature. In this paper, we suggest modifications to the spectral entropy feature extraction approach and compute the entropy contribution from each sub-band to the total entropy of the normalized spectrum. Further, we explore the ideas of overlapping sub-bands and the time derivatives of the spectral entropy feature. The modified feature is robust to additive wide-band noise and performs well at low SNRs. Finally, in the TANDEM framework, we show that the system using combined entropy and PLP (perceptual linear prediction) features works better than the baseline PLP feature for additive wide-band noise at different SNRs.
ISSN:1520-6149
2379-190X
DOI:10.1109/ICASSP.2005.1415098