本文已被:浏览 1816次 下载 1951次
Received:August 11, 2014 Revised:September 05, 2014
Received:August 11, 2014 Revised:September 05, 2014
中文摘要: 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.
Abstract: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%.
keywords: speaker recognition speech enhancement sound source azimuth information time-frequency masking MFCC
文章编号: 中图分类号: 文献标志码:
基金项目:
引用文本:
张凤仪,夏秀渝,冉国敬,何礼,叶于林.多声源环境下的鲁棒说话人识别.计算机系统应用,2015,24(4):32-37
ZHANG Feng-Yi,XIA Xiu-Yu,RAN Guo-Jing,HE Li,YE Yu-Lin.Robust Speaker Recognition in Multi-Source Environment.COMPUTER SYSTEMS APPLICATIONS,2015,24(4):32-37
张凤仪,夏秀渝,冉国敬,何礼,叶于林.多声源环境下的鲁棒说话人识别.计算机系统应用,2015,24(4):32-37
ZHANG Feng-Yi,XIA Xiu-Yu,RAN Guo-Jing,HE Li,YE Yu-Lin.Robust Speaker Recognition in Multi-Source Environment.COMPUTER SYSTEMS APPLICATIONS,2015,24(4):32-37