ISIE 2001. 2001 IEEE International Symposium on Industrial Electronics Proceedings (Cat. No.01TH8570), v. 1, page. 501-504
In this paper, the authors propose an adaptive wavelet threshold for noise cancellation. For this, they use a threshold value which minimizes Bayesian risk. And using entropy, they part the noisy signal into an unvoiced signal section and the other signal section is used to apply each the threshold value for each section. Experimental results show that proposed algorithm is more efficient.