作者投稿和查稿 主编审稿 专家审稿 编委审稿 远程编辑

›› 2005, Vol. 31 ›› Issue (23): 164-165.

• 论文 • 上一篇    下一篇

弹性音节拼接与特征空间轨迹匹配的关键词检测

张传坤;吴亚栋;陆汝占;   

  1. 上海交通大学计算机科学与工程系,上海交通大学计算机科学与工程系,上海交通大学计算机科学与工程系 上海 200030,上海 200030,上海 200030
  • 发布日期:2005-12-15

Keyword Spotting Method of Adjustable Syllable Joining and Feature Space Trace Matching

ZHANG Chuankun,WU Yadong,LU Ruzhan   

  1. (Department of Computer Science and Engineering,Shanghai Jiaotong University,Shanghai 200030)
  • Published:2005-12-15

摘要: 不同的汉语语音关键词识别系统,以关键词为单位建立模板工作量大而且可重用性较差,而以音节为单位可以很好地提高重用性。将弹性音节拼接与具有较高性能的基于特征空间轨迹匹配的关键词检测相结合构建识别系统,实验结果表明该系统具有相当高的识别率。

关键词: 关键词识别, 特征空间轨迹, 弹性音节拼接

Abstract: To establish template for each keyword is laborious and the template isn’t reusable,but choosing syllable as subword can solve the problem.This paper proposes a keyword spotting method based on adjustable syllable joining and feature space trace matching,experiments show that this method has high performance and efficiency.

中图分类号: