Model compensation approach based on nonuniform spectral compression features for noisy speech recognition

Authors:
Geng-Xin Ning;Gang Wei;Kam-Keung Chu
Affiliations:
School of Electronic and Information Engineering, South China University of Technology, Guangzhou, China;School of Electronic and Information Engineering, South China University of Technology, Guangzhou, China;School of Electronic and Information Engineering, South China University of Technology, Guangzhou, China
Venue:
EURASIP Journal on Applied Signal Processing
Year:
2007

Citing 6
Cited 0

Cepstral parameter compensation for HMM recognition in noise

Speech Communication - Special issue on speech processing in adverse conditions
Speech recognition in noisy environments: a survey

Speech Communication
Handbook of Mathematical Functions, With Formulas, Graphs, and Mathematical Tables,

Handbook of Mathematical Functions, With Formulas, Graphs, and Mathematical Tables,
Psychoacoustics: Facts and Models

Psychoacoustics: Facts and Models
A vector Taylor series approach for environment-independent speech recognition

ICASSP '96 Proceedings of the Acoustics, Speech, and Signal Processing, 1996. on Conference Proceedings., 1996 IEEE International Conference - Volume 02
Multichannel direction-independent speech enhancement using spectral amplitude estimation

EURASIP Journal on Applied Signal Processing

Quantified Score

Hi-index	0.00

Visualization

Abstract

This paper presents a novel model compensation (MC) method for the features of mel-frequency cepstral coefficients (MFCCs) with signal-to-noise-ratio- (SNR-) dependent nonuniform spectral compression (SNSC). Though these new MFCCs derived from a SNSC scheme have been shown to be robust features under matched case, they suffer from serious mismatch when the reference models are trained at different SNRs and in different environments. To solve this drawback, a compressed mismatch function is defined for the static observations with nonuniform spectral compression. The means and variances of the static features with spectral compression are derived according to this mismatch function. Experimental results show that the proposed method is able to provide recognition accuracy better than conventional MC methods when using uncompressed features especially at very low SNR under different noises. Moreover, the new compensation method has a computational complexity slightly above that of conventional MC methods.