语音通话质量自动去噪方法研究

Research on auto-denoising method of voice call quality

为了提高语音客户服务质量,通过语音信号的去噪处理,提高语音通话清晰度,提出一种基于小波分析的语音通话信号自动去噪方法。构建客户服务通话语音信号模型,采用调制滤波检测方法进行语音信号初始滤波处理,在滤除语音信号的环境噪声后,采用相关检测方法进行客户服务通话语音信号的波束形成处理,采用正反双曲调频检测方法进行语音信号的谱峰特征检测,根据语音谱峰的频域分布特性进行信号的复包络检测,采用维纳小波尺度分解方法进行语音通话信号的自动去噪处理,提高通话质量。仿真结果表明,采用该方法进行语音信号自动去噪的效果较好,输出信噪比较高,提高了语音通话质量。

In order to improve the quality of voice customer service and improve the clarity of voice call through the denoising processing of speech signal,an automatic denoising method of voice call signal based on wavelet analysis is proposed.The speech signal model of customer service call is constructed,and the initial filtering processing of speech signal is carried out by using modulation filter detection method.After filtering the ambient noise of speech signal,the correlation detection method is used to process the beamforming of the voice signal of the customer service call,and the positive and negative hyperbolic frequency modulation detection method is used to detect the spectrum peak feature of the speech signal.According to the frequency domain distribution characteristics of the speech spectrum peak,the complex envelope detection of the signal is carried out,and the wiener wavelet scale decomposition method is used for the automatic de-noising of the speech call signal to improve the communication quality.The simulation results show that the proposed method is effective and the signal-to-noise ratio(SNR) is high,and the quality of voice communication is improved.