DocumentCode :
2799035
Title :
Efficient VQ-based MMSE estimation for robust speech recognition
Author :
González, José A. ; Peinado, Antonio M. ; Gomez, Angel M. ; Carmona, José L. ; Morales-Cordovilla, Juan A.
Author_Institution :
Dipt. de Teor. de la Senal, Telematica y Comun., Univ. of Granada, Granada, Spain
fYear :
2010
fDate :
14-19 March 2010
Firstpage :
4558
Lastpage :
4561
Abstract :
This paper presents a feature compensation technique based on the minimum mean square error (MMSE) estimation for robust speech recognition. Similarly to other MMSE compensation methods based on stereo data, our approach models the differences between clean and noisy feature spaces, and the resulting MMSE estimate of the clean feature vector is obtained as a piece-wise linear transformation of the noisy one. However, unlike other well-known MMSE techniques such as SPLICE or MEMLIN, which model the feature spaces with GMMs, in our proposal each feature space is characterized by a set of cells obtained by means of VQ quantization. This VQ-based approach allows a very efficient implementation of the MMSE estimator. Also, the possible degradation inherent to any VQ process is overcome by a strategy based on considering different subregions inside each cell and a subregion-based mean and variance compensation. The experimental results show that, along with a a very efficient MMSE estimator, our technique achieves even better recognition accuracies than SPLICE and MEMLIN.
Keywords :
mean square error methods; quantisation (signal); speech recognition; MMSE estimator; VQ quantization; VQ-based MMSE estimation; feature compensation; minimum mean square error estimation; noisy feature space; piece-wise linear transformation; robust speech recognition; stereo data; variance compensation; Acoustic noise; Cepstral analysis; Estimation error; Noise robustness; Piecewise linear techniques; Proposals; Quantization; Speech enhancement; Speech recognition; Working environment noise; MMSE; Speech recognition; noise robustness; stereo data;
fLanguage :
English
Publisher :
ieee
Conference_Titel :
Acoustics Speech and Signal Processing (ICASSP), 2010 IEEE International Conference on
Conference_Location :
Dallas, TX
ISSN :
1520-6149
Print_ISBN :
978-1-4244-4295-9
Electronic_ISBN :
1520-6149
Type :
conf
DOI :
10.1109/ICASSP.2010.5495566
Filename :
5495566
Link To Document :
بازگشت