2006 · 76 citations · 10 references
ComputingPhysicsAcousticsComputational AcousticsPhoneticsAcoustic PhoneticsDigital Signal ProcessingSpectral AnalysisSpectral Data ProcessingSpeech AcousticsSpeech RecognitionCepstral AnalysisCepstral CoefficientsCepstral FeaturesLog CompressionNoise RobustnessNon-stationary NoiseRobust Speech RecognitionSpectral Cues In Speech
It is well known that the peaks in the spectrum of a log Mel-filter bank are important cues in characterizing speech sounds. However, low energy perturbations in the power spectrum may become numerically significant after the log compression. We show that even if the spectral peaks are kept constant, the low energy perturbations in the power spectrum can create huge variations in the cepstral coefficients. We show, both analytically and experimentally, that exponentiating the log Mel-filter bank spectrum before the cepstrum computation can significantly reduce the sensitivity of the cepstra to spurious low energy perturbations. The Mel-cepstrum modulation spectrum (Tyagi, V. et al., Proc. IEEE ASRU, 2003) is computed from the processed cepstra which results in further noise robustness of the composite feature vector. In experiments with speech signals, it is shown that the proposed technique based features yield a significant increase in speech recognition performance in non-stationary noise conditions when compared directly to the MFCC and RASTA-PLP features.
10
Mel-generalized cepstral analysis - a unified approach to speech spectral estimation
Keiichi Tokuda, Takao Kobayashi, Takashi Masuko et al. · 1994 · 237 citations
Spectral subband centroid features for speech recognition
Kuldip K. Paliwal · 2002 · 147 citations · Full text
Computing, Linguistics, Acoustics +14