基于分数阶傅里叶变换的语音消噪

Speech Denoising Based on Fractional-Order Fourier Transform

傅里叶变换是语音信号处理的常用方法,但其存在窗口分辨率的问题,当窗口值大,语音信号的全局特性方便获取,局部信息不精细,当窗口值小,局部信息精度高,全局特性无法完全获取,这是傅里叶变换固有的矛盾。本文采用了分数阶傅里叶变换方法,对有噪音的语音信号先经小波局部消噪处理,利用分数阶傅里叶变换对语音进行了全局消噪,并与循环神经网络消噪方法做了对比实验。实验结果表明,含噪的语音信号经由小波变换的局部和分数阶傅里叶全局消噪的处理后,得到了比循环神经网络方法更好的处理结果。

The Fast Discrete Fourier Transform is commonly used for speech signal processing in frequency domain analysis method.It has the problem of adjusting window size for desired resolution.But the Fractional Fourier Transform can have both time domain and frequency domain processing capabilities.This paper performs global processing by combining Fractional Fourier Transform and wavelet multi-scale local elimination on the speech signal and denoising experiment with RNN neural network.The experimental results show that the speech signal can be processed both locally by wavelet and globally by Fractional Fourier Transform.The combination of the two allows the speech signal to take into account both local and global information in the time and frequency domain.