Non-linear spectral subtraction (NSS) and hidden Markov models for robust speech recognition in car noise environments

Published 1 January 1992

P. Lockwood, J. Boudy, Marie‐Anne Blanchet

DOI10.1109/icassp.1992.225921

Citations53

TL;DR

The authors extend their previous work where a robust hidden Markov model (HMM) training/recognition framework is proposed and several new aspects are introduced: use of enhanced nonlinear spectral subtraction (NSS) schemes, introduction of root-MFCC parameters, and training of HMMs by a dynamic inference scheme (DIHMM).

Abstract

The authors address the problem of speaker-dependent discrete utterance recognition in noise. Special reference is made to the mismatch effects due to the fact that training and testing are carried out in different environments. The authors extend their previous work (Lockwood and Boudy, 1991) where a robust hidden Markov model (HMM) training/recognition framework is proposed. Several new aspects are introduced: use of enhanced nonlinear spectral subtraction (NSS) schemes, introduction of root-MFCC parameters, use of dynamic features, and training of HMMs by a dynamic inference scheme (DIHMM). These enhancements are discussed from tests performed on bandlimited signals (200-3000 Hz). The authors show that these various optimizations allow a rise from 20% to over 99% in performance. A 93% recognition rate is already achievable on raw data using a weighted modified projection and a root-MFCC dynamic representation.>

Keywords

Computer Science

IEEE Transactions on Acoustics Speech and Signal ProcessingComparison of parametric representations for monosyllabic word recognition in continuously spoken sentences

5,322 Citations1980S. Davis, P. Mermelstein

Several parametric representations of the acoustic signal were compared with regard to word recognition performance in a syllable-oriented continuous speech recognition system and the superior performance of the mel-frequency cepstrum coefficients may be attributed to the fact that they better represent the perceptually relevant aspects of the short-term speech spectrum.

The Journal of the Acoustical Society of AmericaPerceptual linear predictive (PLP) analysis of speech

2,539 Citations1990Hynek Heřmanský

A new technique for the analysis of speech, the perceptual linear predictive (PLP) technique, which uses three concepts from the psychophysics of hearing to derive an estimate of the auditory spectrum, and yields a low-dimensional representation of speech.

IEEE Transactions on Acoustics Speech and Signal ProcessingSpeaker-independent isolated word recognition using dynamic features of speech spectrum

741 Citations1986Sadaoki Furui

This paper proposes a new isolated word recognition technique based on a combination of instantaneous and dynamic features of the speech spectrum that is shown to be highly effective in speaker-independent speech recognition.

Speech CommunicationExperiments with a nonlinear spectral subtractor (NSS), Hidden Markov models and the projection, for robust speech recognition in cars

348 Citations1992P. Lockwood, J. Boudy

The performance of an HMM-based recogniser rises from 56% (no compensation) to 98% after speech enhancement and the lower limit of applicability of the projection (low SNR values) can be loosened after combination with NSS.

Computer Speech & LanguageSpeech recognition in adverse environments

198 Citations1991B.-H. Juang

This paper reviews several promising methods that were proposed in the past few years to deal with the problem of automatic speech recognition in an adverse environment, discussing methods or algorithms in six categories: signal enhancement preprocessing; special transducer arrangements; noise masking; stress compensation; robust distortion measures; and novel speech representations.

IEEE Transactions on Acoustics Speech and Signal ProcessingA family of distortion measures based upon projection operation for robust speech recognition

159 Citations1989D. Mansour, B.-H. Juang

It is found that the orientation (or direction) of the cepstral vector is less susceptible to noise perturbation than the vector norm, and a family of distortion measures based on the projection between two cEPstral vectors is proposed, which have the same computational efficiency as the band-pass cepStral distortion measure.

IEEE Transactions on Acoustics Speech and Signal ProcessingSpectral root homomorphic deconvolution system

83 Citations1979Jae S. Lim

Computer Speech & LanguageDTW schemes for continuous speech recognition: a unified view

18 Citations1989Christophe Godin, P. Lockwood

Three dynamic time warping-based search procedures for connected-word recognition, the one-pass, the level-building, and the two-level algorithms, which are shown to be identical when syntactic constraints are applied.

International Conference on Acoustics, Speech, and Signal ProcessingLearning the structure of HMM's through grammatical inference techniques

13 Citations2002Francisco Casacuberta, Enrique Vidal +2 more

Experimental results showing the applicability of the technique described in which all the components of a hidden Markov model are learnt from training speech data are presented.

Application of a weighted projection measure for robust hidden Markov model based speech recognition

13 Citations1991B.A. Carlson, Mark A. Clements

It is shown how the projection measure can be incorporated into a continuous density hidden Markov model system in the form of a weighted measure.