The problem of speech/noise discrimination has become increasingly important as the automatic speech recognition (ASR) system is applied in the real world. Robustness and simplicity are two challenges to the speech/noise discrimination method for an embedded system. The energy-based feature is the most suitable and applicable feature for speech/noise discrimination for embedded ASR system because of effectiveness and simplicity. A new method based on a noise model is proposed to discriminate speech signals from noise signals. The noise model is initialized and then updated according to the signal energy. The experiment shows the effectiveness and robustness of the new method in noisy environments.
A Novel Speech/Noise Discrimination Method for Embedded ASR System
1 Institute of Image Processing & Pattern Recognition, Shanghai Jiaotong University, Shanghai 200030, China
2 Motorola Labs China Research Center, Shanghai 200041, China
EURASIP Journal on Advances in Signal Processing 2004, 2004:951918 doi:10.1155/S111086570440225X
The electronic version of this article is the complete one and can be found online at: http://asp.eurasipjournals.com/content/2004/11/951918
|Received:||30 October 2003|
|Revisions received:||8 February 2004|
|Published:||18 September 2004|
© 2004 Wu et al.