基音同步帧长特征在英语词重音检测中的应用  

Application of pitch synchronization dynamic frame-length features in English lexical stress detection

作　　者：陈楠[1] 贺前华[1] 王伟凝[1] 陈荣研[1] 

机构地区：[1]华南理工大学电子与信息学院,广州510640

出　　处：《计算机应用》2008年第6期1533-1536,共4页journal of Computer Applications

基　　金：国家自然科学基金资助项目(6057214160602014)

摘　　要：对于英语等"重音节拍语言",重音是一个非常重要的韵律学特征。针对传统特征提取中固定帧长方式存在的缺点,使用基音同步帧特征分析方法,提出了基于动态帧长的基音同步能量和基音同步峰值特征。在使用新特征对英语连续语音进行词重音检测时发现,联合使用新特征与传统特征,可使误识率下降6.65%。Lexical stress is an important prosodic feature, especially for stress-timed language such as English. To overcome the defects of fixed frame-length features, pitch synchronization feature analysis method was proposed while Pitch Synchronization Energy （PSE） and Pitch Synchronization Peak （PSP） features were defined and extracted. Their contributions, along with traditional features and their combinations, to English lexical stress detection were evaluated with ISLE database. Experimental results show that the combination of new feature and traditional features demonstrates a 6.65% error rate reduction compared with using traditional ones.

关 键 词：重音检测 动态帧长 基音同步 

分 类 号：TP391.4[自动化与计算机技术—计算机应用技术] TN912.34[自动化与计算机技术—计算机科学与技术]