摘要
提出了一种基于小波包变换的自适应阈值语音去噪净化方法。采用双正交小波包分解,克服了传统的正交小波变换的缺陷。采用自适应阈值方法,通过最佳阈值的估计,在去噪的同时进一步提高信噪比。仿真实验结果表明,方法能在有效去除噪声的过程中保证相位不失真和较好保留语音细节,语音净化效果明显优于其他方法。
An adaptive threshold method based on wavelet packet transform of speech denoising and purification is proposed. Biorthogonal wavelet packet decomposition overcomes the weakness of the conventional orthogonal wavelet transform. Adaptive threshold method, by the optimal estimation of the threshold, can increase the SNR as well as denoising. The simulation results show that the new method can not only ensure non-distortion, but also hold the main details better than the conventional ones during removing the strong noise more effectively.
出处
《电声技术》
2009年第10期69-72,共4页
Audio Engineering
关键词
小波包变换
自适应阈值法
语音净化
wavelet packet transform
adaptive threshold method
speech purification
