多声源环境下的鲁棒说话人识别  

Robust Speaker Recognition in Multi-Source Environment

作　　者：张凤仪[1] 夏秀渝[1] 冉国敬 何礼[1] 叶于林 

机构地区：[1]四川大学电子信息学院,成都610065

出　　处：《计算机系统应用》2015年第4期32-37,共6页Computer Systems & Applications

摘　　要：针对多声源干扰环境下说话人识别系统性能急剧下降的问题,提出一种提取目标语音的前端处理方法,该方法依据独立语音时频域的近似稀疏性,基于目标语音方位信息采用非线性时频掩蔽方法提取目标语音.建立了基于梅尔倒谱系数(MFCC)的高斯混合模型(GMM)说话人识别系统.仿真实验证明,该方法能有效提取目标语音,提高说话人识别系统的鲁棒性.该文多声源干扰仿真实验条件下,说话人识别系统的识别率平均提高了25%左右.The Speaker Recognition System is significantly affected by the Multi-Sound sources problem. In order to overcome this problem, a target sound extraction algorithm named time-frequency masking is proposed. The proposed algorithm is based on the sound source azimuth information and the approximate sparse nature of sound. A Mel-frequency cepstral coefficient （MFCC） based Gaussian mixture model （GMM） speaker recognition system is presented to improve the recognition robustness. The proposed algorithm has been tested on the simulated data through a number of experiments which shows the efficiency and robustness of the proposed algorithm. In the Multi-Sound sources environment, the recognition rate of the proposed algorithm can be improved by about 25%.

关 键 词：说话人识别 语音增强 方位信息 时频掩蔽 MFCC参数 

分 类 号：TN912.34[电子电信—通信与信息系统]