| 基于模式识别的语音分类方法 |
| |
| 引用本文： | 朱琦,酆广增,肖海勇.基于模式识别的语音分类方法[J].南京邮电大学学报(自然科学版),2000,20(4):29-33. |
| |
| 作者姓名： | 朱琦 酆广增 肖海勇 |
| |
| 作者单位： | 南京邮电学院,通信工程系,江苏,南京,210003 |
| |
| 摘 要： | 对语音的浊/清分类方法进行了讨论,并着重研究了基于多个特征参数、应用模式识别理论的语音分类系统.实验表明,采用Fisher分类器实现的语音分类方法,对训练内语音分类准确率可达98.6%,而训练外语音分类准确率也可达96.7%.
|
| 关 键 词： | 语音数据处理 模式识别 分类方法 |
| 文章编号： | 1000-1972(2000)04-0029-05 |
| 修稿时间： | 1999年9月2日 |
A Pattern Recognition Decision Method for the Voiced/Unvoiced Classification of Speech |
| |
| ZHU Qi,FENG Guang-zeng,XIAO Hai-yong.A Pattern Recognition Decision Method for the Voiced/Unvoiced Classification of Speech[J].Journal of Nanjing University of Posts and Telecommunications,2000,20(4):29-33. |
| |
| Authors: | ZHU Qi FENG Guang-zeng XIAO Hai-yong |
| |
| Abstract: | Some methods for the classification of voiced/unvoiced speech are explored in this paper,and the focus is on the decision-making process,which can be viewed as a pattern recognition problem.Experiments demonstrate that the classifier based on the Fisher rule can achieve 98.6 percent classification accuracy on speaker dependent tests,and 96.7 percent accuracy on speaker independent tests. |
| |
| Keywords: | Speech data processing Pattern recognition Classification syste |
| 本文献已被 CNKI 万方数据 等数据库收录！ |
|
