基于DTW算法的语音评分系统的设计与实现
首发时间:2011-01-12
摘要:在现代英语教学中,计算机辅助教学正日益普及,对英语发音的准确度的要求也不断提高。借助于语音识别技术,实现学生发音效果的实时反馈,已经成为当前研究的一个热点。本文结合音频编程,DTW算法以及Access数据库,设计了一套语音评分系统。借助音频编程函数,实现语音数据的录制,播放以及波形显示功能;通过DTW模板训练,实现基于MFCC特征参数的模板训练以及训练语音和标准语音匹配距离的计算;通过Access数据库操作,实现模板参数的存储和读取。最后,根据评分准则函数,实现语音评分。
For information in English, please click here
The Design and Realize of Speech Scoring System Based on DTW Algorithm
Abstract:In modern English education, computer-assisted language learning has become more and more popular and the demand for English pronunciation accuracy has become more and more high. Now, it becomes the research focus to feedback the effect of student's pronunciation in time by speech recognition technology. A speech scoring system, using audio programming, DTW algorithm and Access database, is designed in this paper. The audio programming module realizes the speech data recording ,playing and waveform displaying; the DTW template training module realizes the template training based on MFCC parameters and distance computing between training data and standard data; the Access module realizes the template parameters reading and saving. Finally, the score is given according to the scoring rules.
Keywords: speech scoring system DTW algorithm template training
基金:
论文图表:
引用
No.4402886561389129****
同行评议
共计0人参与
勘误表
基于DTW算法的语音评分系统的设计与实现
评论
全部评论0/1000