Real time voice denoising method and device

Abstract

The invention provides a real time voice denoising method and device; the method comprises the following steps: generating a frequency domain zone noise voice signal according to a voice input received by a voice receiver; calculating a logarithm spectrum posterior signal to noise ratio according to the frequency domain zone noise voice signal, wherein the logarithm spectrum posterior signal to noise ratio refers to a ratio between logarithm of a power spectrum of a present frame frequency domain zone noise voice signal and a logarithm of a previous frame noise power estimation value; obtaining a noise power spectrum estimation value according to the logarithm spectrum posterior signal to noise ratio and based on a weight noise estimation algorithm; generating a Wiener filtering gain function according to the noise power spectrum estimation value, and filtering the frequency domain zone noise voice signal according to the gain function, thus generating a frequency domain denoising voice signal; generating a time domain denoising voice signal according to the frequency domain denoising voice signal, and the time domain denoising voice signal is further processed by the voice receiver. Correspondingly, the invention also provides the real time voice denoising device.

Description

technical field [0001] The invention relates to the field of voice digital processing, in particular to a method and device for real-time voice denoising. Background technique [0002] In terms of noise suppression, the Wiener filter algorithm has always been the most important and the most effective estimation algorithm, and is widely used in various fields such as image, video, and voice processing. Among them, in terms of speech denoising, there are currently many speech denoising methods based on Wiener filtering. However, these methods generally cannot be well applied to voice receiving devices with limited processing capabilities, such as smart mobile terminals. Taking the smart mobile terminal as an example, the application limitations of this method are as follows: First, the existing voice denoising method is not fast enough to track the noise, and the complexity of the method is relatively high, so it does not meet the real-time requirements of the smart mobile te...

AI Technical Summary

Problems solved by technology

However, these methods usually cannot be well applied to voice receiving devices with limited processing capabilities, such as smart mobile terminals

Taking the smart mobile terminal as an example, the application limitations of this method are as follows: First, the existing voice denoising method is not fast enough to track the noise, and the complexity of the method is relatively high, so it does not meet the real-time requirements of the smart mobile terminal. Computational requirements; second, when estimating real-time noise, the existing practice is usually to use the initial frame of the noisy speech signal as the initial noise. Tracking noise, which in turn causes the processed sound to be distorted during this time

Although this period of time is usually relatively short, users using the smart mobile terminal will still feel the distortion of the sound very obviously, which will have a certain impact on the user experience

In addition, the current speech denoising method based on Wiener filtering is not perfect in distinguishing weak speech from noise, so it is easy to cause distortion of weak speech

Images