噪声环境中的双声道音频回波抵消模型构建

Construction of a dual channel audio echo cancellation model in noisy environments

为提升远端说话人员的分辨率,保证真实空间化听觉感受效果,构建一种噪声环境中的双声道音频回波抵消模型。该模型以子带自适应滤波器组为核心分解音频信号,使其形成若干个子频带信号;在子带域内计算噪声干扰阈值,依据阈值的计算结果并结合人耳听觉特性,叠加独立分布白噪声,实现各个子带的回波抵消。在此基础上,引入步长补偿因子和反馈控制理念改进最小均方算法,优化滤波器组参数,提升双声道音频回波抵消效果。测试结果表明:所设计模型可有效处理音频信号中的噪声信号,抵消后的音频信号信噪比均在3.5 dB以上,音频信号的振幅误差均在0.2 V以下,能够提升远端人员说话的分辨率,保证音频质量。

In order to improve the resolution of remote speakers and ensure a realistic spatial auditory experience,a dual channel audio echo cancellation model in a noisy environment is constructed.This model can decompose audio signals with subband adaptive filter banks as the core,forming several subband signals.The noise interference threshold in the subband domain is calculated,and the independently distributed white noise is superimposed according to the calculation results of the threshold and combined with the auditory characteristics of human ear,so as to achieve echo cancellation for each subband.On this basis,the concept of step compensation factor and feedback control is introduced to improve the minimum mean square algorithm to optimize filter bank parameters and enhance the cancellation effect of dual channel audio echoes.The testing results show that the designed model can effectively process noise signals in audio signals,and the signal-to-noise ratio results of the cancelled audio signals are all above 3.5 dB.The amplitude errors of the audio signals are all below 0.2 V,which can improve the resolution of remote personnel speaking and ensure audio quality.