首页> 外文会议>International Conference on Intelligent Computing and Integrated Systems >Application of probability density estimation criteria in speaker recognition
【24h】

Application of probability density estimation criteria in speaker recognition

机译:概率密度估计准则在说话人识别中的应用

获取原文

摘要

With the rapid development of computer technology, speaker recognition has been widely used and researched. In this paper, a traditional code book based on VQ algorithm was improved by applying in probability density estimation criteria method. Through the method, the problem of easily falling into local optimum existing in traditional LBG codebook can be well solved, and controlling operation can be easily achieved. Experimental results show that improved codebook algorithm has less distortion than traditional LBG algorithm.
机译:随着计算机技术的飞速发展,说话人识别得到了广泛的应用和研究。本文采用概率密度估计准则方法,对基于VQ算法的传统密码本进行了改进。通过该方法,可以很好地解决传统的LBG码本中容易陷入局部最优的问题,并且可以容易地实现控制操作。实验结果表明,改进的码本算法比传统的LBG算法具有更少的失真。

著录项

相似文献

  • 外文文献
  • 中文文献
  • 专利
获取原文

客服邮箱:kefu@zhangqiaokeyan.com

京公网安备:11010802029741号 ICP备案号:京ICP备15016152号-6 六维联合信息科技 (北京) 有限公司©版权所有
  • 客服微信

  • 服务号