Title :
Combining MMSE enhancement with LA model adaptation for robust automatic speech recognition
Author :
Ding, Pei ; Cao, Zhigang
Author_Institution :
Dept. of Electron. Eng., Tsinghua Univ., Beijing, China
fDate :
4/12/2001 12:00:00 AM
Abstract :
A new scheme for robust automatic speech recognition is presented. The scheme combines minimum mean square error (MMSE)-based enhancement with Log-Add (LA) model adaptation. Experimental evaluations show that the proposed MMSE-LA scheme can achieve significant improvement in recognition across a wide range of signal-to-noise ratios, especially in very noisy environments
Keywords :
adaptive signal processing; least mean squares methods; noise; speech recognition; LA model adaptation; MMSE enhancement; SNR; log-add model adaptation; minimum mean square error based enhancement; noisy environments; recognition improvement; robust automatic speech recognition; signal-to-noise ratios;
Journal_Title :
Electronics Letters
DOI :
10.1049/el:20010341
