【摘要】 语音增强技术是解决噪声干扰的一个有效途径.通过分析人耳听觉掩蔽特性,提出一种改进的语音增强算法.该算法将人耳听觉掩蔽特性与多频带谱相减结合,首先对带噪语音进行多频带谱相减,然后从感知的角度,利用估计的语音信号计算掩蔽阈值,合理地调整谱减参数.仿真结果表明,与传统算法相比,该算法在低信噪比情况下能很好地抑制背景噪声和残余的音乐噪声,同时可保持较好的语音质量.更多还原

【Abstract】 The speech enhancement is a powerful treatment technology for dealing with the noise pollution.According to the human auditory masking properties,the paper proposes an improved method of speech enhancement.In this method,it combines masking properties of human auditory system with multi-band spectral subtraction algorithm.Firstly,this method performs multi-band spectral subtraction,then,the estimated results can be used to compute the masking threshold of the speech,after that,the coefficients are adjusted according to perception.The simulation results show its excellent performance in background noise and musical noise in low SNR condition while keeping very low audible distortion.更多还原