面向双语教学的维吾尔语发音质量自动评估  

Bilingual teaching-facing automatic assessment for Uygur pronunciation

作　　者：董滨[1] 丁铭[1] 王磊[2] 颜永红[1] 

机构地区：[1]中国科学院声学研究所语言声学与内容理解重点实验室,北京100190 [2]中国科学院新疆理化技术研究所,乌鲁木齐830011

出　　处：《中国科学：信息科学》2015年第10期1328-1340,共13页Scientia Sinica(Informationis)

基　　金：新疆维吾尔自治区科技重大专项(批准号:201230118-3)资助项目

摘　　要：推行新疆地区的双语教学是国家推进民族交流的重要举措,对维语进行自动发音质量评估能够大大提高双语教学的效率.然而维吾尔语作为黏着语,其特殊的构词方法造成大量无法被收入词表的集外词存在,使得基于传统语音识别系统的发音评估方法难以应用.为了实现高效的维语发音质量评估,本文在研究分析了维语的发音规则和发音习惯后,决定采用子词作为基本的识别单元;在原有发音质量自动评估系统上,改用基于双层词法分析的维语子词切分器;同时引入音素解码器计算后验概率的分母,让系统可以在子词级别直接计算置信度.经过实验数据的验证,基于子词的维语评估系统的性能要优于基于音素的系统.Pursuing Chinese and Uygur bilingual teaching in Xin Jiang Province is an important government policy for improving communication between different language speakers. An automatic assessment system can yield many advantages in such bilingual teaching. However, as an agglutinative language, Uygur＇s particular way of word-building results in leaving many words out of the lexicon. Therefore, performance of the old pronunciation assessment system, which is based on a traditional ASR system, is poor. For building a high-performance system,we decide to use subwords as basic recognition units after analyzing the rules and habits of Uygur pronunciation.Experimental results indicate that the accuracy of the subword-based system is greatly improved by the implementation of machine segmentation based on double-level morphology analysis of Uygur subwords, calculating the posterior probability＇s denominator with a phoneme decoder, and calculating confidence at the subword level.

关 键 词：双语教学 黏着语 发音评估 子词 后验概率 置信度 

分 类 号：TN912.3[电子电信—通信与信息系统]