Exploiting articulatory features for pitch accent detection  

作　　者：Junhong ZHAO Ji XU Wei-qiang ZHANG Hua YUAN Jia LIU Shanhong XIA 

机构地区：[1]State Key Laboratory of Transducer Technology,Institute of Electronics,Chinese Academy of Sciences [2]University of Chinese Academy of Sciences [3]National Laboratory for Information Science and Technology,Department of Electronic Engineering,Tsinghua University

出　　处：《Journal of Zhejiang University-Science C(Computers and Electronics)》2013年第11期835-844,共10页浙江大学学报C辑（计算机与电子（英文版）

基　　金：Project(Nos.61370034,61273268,and 61005019) supported by the National Natural Science Foundation of China

摘　　要：Articulatory features describe how articulators are involved in making sounds.Speakers often use a more exaggerated way to pronounce accented phonemes,so articulatory features can be helpful in pitch accent detection.Instead of using the actual articulatory features obtained by direct measurement of articulators,we use the posterior probabilities produced by multi-layer perceptrons(MLPs) as articulatory features.The inputs of MLPs are frame-level acoustic features pre-processed using the split temporal context-2(STC-2) approach.The outputs are the posterior probabilities of a set of articulatory attributes.These posterior probabilities are averaged piecewise within the range of syllables and eventually act as syllable-level articulatory features.This work is the first to introduce articulatory features into pitch accent detection.Using the articulatory features extracted in this way,together with other traditional acoustic features,can improve the accuracy of pitch accent detection by about 2%.Articulatory features describe how articulators are involved in making sounds. Speakers often use a more exaggerated way to pronounce accented phonemes, so articulatory features can be helpful in pitch accent detection. Instead of using the actual articulatory features obtained by direct measurement of articulators, we use the posterior probabilities produced by multi-layer perceptrons （MLPs） as articulatory features. The inputs of MLPs are frame-level acoustic features pre-processed using the split temporal context-2 （STC-2） approach. The outputs are the posterior probabilities of a set of articulatory attributes. These posterior probabilities are averaged piecewise within the range of syllables and eventually act as syllable-level articulatory features. This work is the first to introduce articulatory features into pitch accent detection. Using the articulatory features extracted in this way, together with other traditional acoustic features, can improve the accuracy of pitch accent detection by about 2%.

关 键 词：Articulatory features Pitch accent detection PROSODY Computer-aided language learning(CALL) Multi-layer perceptron(MLP) 

分 类 号：TN912.34[电子电信—通信与信息系统]