噪声环境中基于VQ说话人识别

Analysis of VQ to Speaker Recognition in Noisy Environment

噪声环境下,为了提高说话人识别系统的鲁棒性,需要对系统进行各种抗噪声处理。采用梅尔频率倒谱系数作为语音的特征参数,矢量量化方法进行模式匹配,将改进的基于听觉掩蔽效应的语音增强器作为预处理器,对语音信号首先进行降噪处理。语音增强器实验结果表明,经过降噪处理后提高了输入信号的信噪比,减少了语音失真,同时很好地抑制了背景噪声和残余音乐噪声。将经过降噪处理的语音信号送入说话人识别系统,提高了系统的识别性能。

In a noisy environment, in order to improve robustness of speaker recognition system, a variety of anti - noise processing are required for system. Taking MFCC as the voice feature parameter, and VQ as the model matching method. Speech enhancement method based on improved masking properties of the human auditory system is used to reduce the white noise in the front -end. Experimental results show that the improved method leads to better signal to noise ratio, significant reduction of background noise,and the performance of speaker recognition system is proved with the processed speech signal in the noise environment.