首页 | 本学科首页   官方微博 | 高级检索  
     

语音识别特征参数选择方法研究
引用本文:杨大利,徐明星,吴文虎.语音识别特征参数选择方法研究[J].计算机研究与发展,2003,40(7):963-969.
作者姓名:杨大利  徐明星  吴文虎
作者单位:清华大学计算机科学与技术系,北京,100084
基金项目:国家自然科学基金 ( 198710 45 )
摘    要:语音识别系统的性能与识别器所用的特征参数密切相关,用正交实验设计进行特征参数选择,提出一种系统性的实用的特征参数选择方法.研究了特征参数排列顺序对识别系统最终识别率的影响,讨论了这种特征参数选择方法的可扩展性.与目前常用识别器相比,新方法的词错误率下降了37%.实验结果表明,正交实验设计用于特征参数选择是有效的,特征参数排列顺序对识别率的影响可以忽略.

关 键 词:语音识别  特征选择  正交实验  因子顺序

Study of Feature Selection for Speech Recognition
YANG Da Li,XU Ming Xing,and WU Wen Hu.Study of Feature Selection for Speech Recognition[J].Journal of Computer Research and Development,2003,40(7):963-969.
Authors:YANG Da Li  XU Ming Xing  and WU Wen Hu
Abstract:As well known, the performance of the recognizer depends on the features of recognizer In this paper, a systematic and practical method of the features selection based on the orthogonal design is proposed The influence of the features' order on the performance of the recognition system and the generalization of the features selection method are studied The word error rate of the method is reduced 37% when compared with that of the traditional recognizer The experiment results show that the application of orthogonal experiment design is effective in feature selection, and the influence of the features' order on the performance of the recognition system can be ignored
Keywords:speech recognition  feature selection  orthogonal design  factors' order
本文献已被 CNKI 维普 万方数据 等数据库收录!
设为首页 | 免责声明 | 关于勤云 | 加入收藏

Copyright©北京勤云科技发展有限公司  京ICP备09084417号