摘要
人耳对于声音信号的分辨率是非线性的,而仿生小波变换正是由非线性的耳蜗听觉模型发展而来。文章提出一种基于仿生小波变换的语音信号去噪方法,同时针对软硬阈值函数的不足,采用一种修正的硬阈值函数进行处理。仿真结果表明,该方法具有较好的增强效果。
The frequency resolution of human ear has non-linear characteristics,while BWT(Bionic Wavelet Transform)derives from a nonlinear auditory model of the cochlea.In this paper,a algorithm is proposed for speech de-noising using Bionic Wavelet Transform, some modification is given to hard threshold approach for its drawback.Simulation results indicate that this new method has improved the effect of denoising.
出处
《电子技术（上海）》
2008年第2期59-61,共3页
Electronic Technology
关键词
仿生小波变换
语音增强
阈值函数
bionic wavelet Transform
speech enhancement
threshold function
