汉语普通话水平测试中儿化音的自动检测与评价  

Automatic detection and evaluation of the Erhua in Putonghua proficiency test

作　　者：张珑[1,2] 李海峰[1] 马琳[1] 王建华[2] 

机构地区：[1]哈尔滨工业大学计算机科学与技术学院,哈尔滨150001 [2]哈尔滨师范大学计算机科学与信息工程学院,哈尔滨150025

出　　处：《声学学报》2014年第5期639-646,共8页Acta Acustica

基　　金：国家自然科学基金(61171186);黑龙江省自然科学基金(F201321);黑龙江省应用技术研究与开发计划项目(GZ13A003)资助

摘　　要：提出一种汉语普通话水平测试中儿化音的自动检测与评价方法。在现有计算机辅助发音评测系统的框架下,深入分析儿化音的发音规律和声学特性,将儿化音的检测与评价转化成典型的分类问题进行处理。经过挑选多个有代表性的声学特征,并尝试多种不同的分类算法,结果表明,集成分类回归树(Boosting CART)强化分类模型,能充分利用儿化音的各种声学特征,分类正确率达到92.41%。通过对声学特征组的进一步分析,发现共振峰、发音置信度、时长是表达儿化音的最重要线索,利用这些线索能有效地实现对儿化音的自动检测与评价。An automatic detection and evaluation method of the ＂Erhua＂ （also called ＂r-retrofiexion＂ or ＂retroflex suffixation＂） in Putonghua proficiency test （PSC） is proposed. Based on the framework of the computer aided pronun- ciation evaluation system, this paper makes an in-depth analysis of phonologic rules and acoustic characteristic of the Erhua, solves the detection and evaluation of the Erhua as a typical classification problem. And then more representa- tive acoustic features are selected and a variety of different classification algorithms are used. The results showed that the boosting classification and regression tree （CART） can make full use of the characteristics of the Erhua, and the classification accuracy is 92.41%. Based on the further analysis of the acoustic feature group, we found the formant, pronunciation confidence and duration are the most important clues of the Erhua, and using these clues can effectively realize the automatic detection and evaluation of the Erhua.

关 键 词：普通话水平测试 自动检测 评价方法 汉语 声学特征 分类问题 计算机辅助 分类回归树 

分 类 号：TN912.3[电子电信—通信与信息系统]