基于听觉模型的特征在英语重音检测中的应用  

Application of Auditory Model-based Feature in English Lexical Stress Detection

作　　者：陈楠[1] 贺前华[1] 李韬[1] 

机构地区：[1]华南理工大学电子与信息学院,广州510641

出　　处：《计算机工程》2009年第8期26-27,30,共3页Computer Engineering

基　　金：国家自然科学基金资助项目(60572141;60602014)

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。从听觉模型的角度出发,利用基音同步幅度峰值特征能同时表征瞬时频率和强度信息的特点进行重音检测。使用基音同步幅度峰值特征以及与传统特征的组合对英语连续语音的试验结果表明,新特征能使系统误识率降低1.5%。Stress is an important prosodic feature for stress-timed language such as English, This paper presents a new approach using auditory model-based Pitch Synchronous Peak Amplitude（PSPA） feature, which incorporates frequency and intensity information in English lexical stress detection. PSPA feature, along with traditional features and their combinations to English lexical stress detection are evaluated with ISLE database. Experimental results show that the combination of new feature and traditional features demonstrates a 1.5% error rate reduction.

关 键 词：重音检测 听觉模型 基音同步幅度峰值 

分 类 号：TN912.34[电子电信—通信与信息系统]