Modulation spectrum power-law expansion for robust speech recognition

In this paper, we present a novel approach to enhancing the speech features in the modulation spectrum for better recognition performance in noise-corrupted environments. In the presented approach, termed modulation spectrum power-law expansion (MSPLE), the speech feature temporal stream is first pre-processed by some statistics compensation technique, such as mean and variance normalization (MVN), cepstral gain normalization (CGN) and MVN plus ARMA filtering (MVA), and then the magnitude part of the modulation spectrum (Fourier transform) for the feature stream is raised to a power (exponentiated). We find that MSPLE can highlight the speech components and reduce the noise distortion existing in the statistics-compensated speech features. With the Aurora-2 digit database task, experimental results reveal that the above process can consistently achieve very promising recognition accuracy under a wide range of noise-corrupted environments. MSPLE operated on MVN-preprocessed features brings about 55% in error rate reduction relative to the MFCC baseline and significantly outperforms the single MVN. Furthermore, performing MSPLE on the lower sub-band modulation spectra gives the results very close to those from the full-band modulation spectra updated by MSPLE, indicating that a less-complicated MSPLE suffices to produce noise-robust speech features.