A family of MLP based nonlinear spectral estimators for noise reduction

In this paper we present a family of nonlinear spectral estimators for noise reduction which are approximated and implemented by a multilayer perceptron neural network. The estimators are approximations of the true minimum mean square error estimator in the logarithmic or a related perceptual domain. Training data for the neural networks is generated from relevant statistical speech and noise models. One single estimator network is generated for all frequency channels. Parameters describing both the noise and speech distribution are estimated on line and provided as extra inputs to the neural net. Including these parameters significantly improves performance over standard spectral estimators which are based on a global speech model and a noise model described by a single parameter, the noise mean.<<ETX>>

[1]  Richard M. Schwartz,et al.  Enhancement of speech corrupted by acoustic noise , 1979, ICASSP.

[2]  D. van Compernolle.  Spectral estimation using a log-distance error criterion applied to speech recognition , 1989, ICASSP.

[3]  Roger K. Moore,et al.  Hidden Markov model decomposition of speech and noise , 1990, International Conference on Acoustics, Speech, and Signal Processing.

[4]  Dirk Van Compernolle.  Noise adaptation in a hidden Markov model speech recognition system , 1989 .

[5]  Kurt Hornik,et al.  Multilayer feedforward networks are universal approximators , 1989, Neural Networks.

[6]  Jérôme Boudy,et al.  Experiments with a nonlinear spectral subtractor (NSS), Hidden Markov models and the projection, for robust speech recognition in cars , 1991, Speech Commun..

[7]  Steven F. Boll,et al.  Optimal estimators for spectral restoration of noisy speech , 1984, ICASSP.

[8]  Mitch Weintraub,et al.  Energy conditioned spectral estimation for recognition of noisy speech , 1993, IEEE Trans. Speech Audio Process..

[9]  Fei Xie,et al.  Speech enhancement by nonlinear spectral estimation - a unifying approach , 1993, EUROSPEECH.

[10]  David Malah,et al.  Speech enhancement using a minimum mean-square error log-spectral amplitude estimator , 1984, IEEE Trans. Acoust. Speech Signal Process..

[11]  Yariv Ephraim,et al.  Speech enhancement based upon hidden Markov modeling , 1989, International Conference on Acoustics, Speech, and Signal Processing,.

[12]  S. Tamura,et al.  An analysis of a noise reduction neural network , 1989, International Conference on Acoustics, Speech, and Signal Processing,.

[13]  Harald Eckhardt,et al.  Combination of distortion-robust feature extraction and neural noise reduction for ASR , 1993, EUROSPEECH.