一种新的听觉特征提取算法研究

New algorithm for auditory feature extraction

人类听觉系统具有显著的分析和识别信号的能力,一个好的听觉特征提取算法能提高说话人识别系统的性能。文中提出的算法从模拟人类听觉角度出发,首先采用伽马通(Gammatone)滤波器组代替传统的三角滤波器组来模拟人耳耳蜗的听觉模型,然后引入等响曲线进行等响预加重,初步提取出GFCC,最后经过半升正弦函数倒谱提升技术提取说话人语音信号的特征参数NewGFCC。实验表明,该算法在说话人识别系统中具有较高的识别率和鲁棒性。

The human auditory system has signal analysis and recognition ability. Thus, a good auditory feature extraction algorithm can improve the performance of the voiceprint recognition system. The algorithm is based from the simulation of human auditory perspective. Firstly, the gamma through （Gammatone） filter bank is used instead of the traditional triangular filter wave device to simulate the human cochlea of auditory model. Then, the equal loudness curve is introduced to make equal loudness pre-em- phasis, the GFCC is preliminarily extracted. Finally, the signal characteristic parameter NewGFCC of speaker＇ s voice is extracted by a half-sine function cepstrum lifting technology. Experimental results show that the algorithm has higher recognition and robustbess in the speaker recognition system.