Comparison and combination of features in a hybrid HMM/MLP and a HMM/GMM speech recognition system

Recently, the advantages of the spectral parameters obtained by frequency filtering (FF) of the logarithmic filter-bank energies (logFBEs) have been reported. These parameters, which are frequency derivatives of the logFBEs, lie in the frequency domain, and have shown good recognition performance wi...

Full description

Saved in:
Bibliographic Details
Published in:IEEE transactions on speech and audio processing 2005-01, Vol.13 (1), p.14-22
Main Authors: Pujol, P., Pol, S., Nadeu, C., Hagen, A., Bourlard, H.
Format: Article
Language:English
Subjects:
Applied sciences
Cepstral analysis
Cleaning
Exact sciences and technology
Feature recognition
Filtering
Filtration
Frequency domain analysis
Frequency filtering (FF)
Hidden Markov models
Information, signal and communications theory
Mathematical models
Mel frequency cepstral coefficient
MLP
multistream
Nonlinear filters
Predictive models
product rule
Recognition
relative spectra (Rasta)
Robustness
Signal processing
Speech
Speech processing
Speech recognition
Studies
System testing
Telecommunications and information theory
Citations: Items that this one cites
Items that cite this one
Online Access:Get full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:Recently, the advantages of the spectral parameters obtained by frequency filtering (FF) of the logarithmic filter-bank energies (logFBEs) have been reported. These parameters, which are frequency derivatives of the logFBEs, lie in the frequency domain, and have shown good recognition performance with respect to the conventional mel-frequency cepstral coefficients (MFCCs) for hidden Markov models (HMM) based systems. In this paper, the FF features are first compared with the MFCCs and the relative spectral perceptual linear prediction (Rasta-PLP) features using both a hybrid HMM/MLP and a usual HMM/Gaussian mixture models (HMM/GMM) based recognition system, for both clean and noisy speech. Taking advantage of the ability of the hybrid system to deal with correlated features, the inclusion of both the frequency second-derivatives and the raw logFBEs as additional features is proposed and tested. Moreover, the robustness of these features in noisy conditions is enhanced by combining the FF technique with the Rasta temporal filtering approach. Finally, a study of the FF features in the framework of multistream processing is presented. The best recognition results for both clean and noisy speech are obtained from the multistream combination of the J-Rasta-PLP features and the FF features.
ISSN:1063-6676
2329-9290
1558-2353
2329-9304
DOI:10.1109/TSA.2004.834466