A New Speech Denoising Algorithm Based on Bilateral Filtering and Wavelet Transform

摘要

A speech denoising algorithm based on bilateral filtering and wavelet transform is proposed in this paper in order to solve the noise problem in speech processing. After the discrete wavelet transform, the low-frequency part of the speech signal is filtered by one-dimensional bilateral filtering. Then the soft thresholding wavelet denoising method is performed on the high-frequency part. The experiments show that this method can effectively reduce the noise while fully reserving the signal detail.