2[1]Boll S.Suppression of acoustic noise in speech using spectral subtraction[J].IEEE and Signal processing,1979,27 (2):113 -120.
3[3]Furuis,Itoh D.Neural-network-based HMM adaptation for noisy speech[C].Processing of 2001 IEEE International conference on ASSP,2001:365-368.
4[5]Mallat S.A theory of multiresolution signal decomposition:The wavelet transform[J].IEEE Trans,1989,11 (7):647-693.
5[6]杨行峻,迟惠生.语音数字信号处理[M].北京:电子工业出版社,2000:129-130.
6Gales M J F, Young S J. Robust Continuous Speech Recognition Using Parallel Model Combination [J]. IEEE Transactions on Speech and Audio Processing, 1996, 4 (5):352 - 359.
7Yao K S,Bertram E Shi,Fung P,et al. Log- Add Approximation On - line Noise Compensation Method Adopting Robust Decision Rules for Robust Digits Recognition [J]. Chinese Journal of Electronics, 2000,9(3) :278 - 283.
