英语口语测试系统发音错误智能检测技术研究  

Research on the intelligent detection technology of pronunciation errors in spoken english test system

作　　者：谢雪梅[1] XIE Xuemei(Yulin College,Yulin Shaanxi,719000)

机构地区：[1]榆林学院

出　　处：《自动化与仪器仪表》2018年第12期58-61,共4页Automation & Instrumentation

基　　金：榆林学院教学改革项目“教育生态学视域下大学英语生态课堂模式建构”(JG1706)

摘　　要：在英语口语测试系统设计中,为提高口语发音错误的智能检测能力,提出一种基于语音信号频谱检测技术的英语口语测试系统发音错误智能检测方法。采集英语口语测试系统发音的语音信号,对语音信号进行匹配滤波处理,去除干扰噪声,对提纯的英语语音信号进行自适应波束聚焦处理,增强有用语音信号的强度,采用相关频谱检测方法,提取语音口语测试系统的发音语音信号的频谱,根据频谱差异性实现语音发音错误检测。仿真结果表明,采用该方法进行英语口语测试系统发音错误智能检测性较好,检测准确率较高,具有很好的抗干扰能力和纠错能力。In the design of oral English testing system,in order to improve the ability of intelligent detection of spoken English pronunciation errors,a new method based on speech signal spectrum detection technology is proposed to detect pronunciation errors in oral English testing system.The speech signal of English spoken language test system is collected,the speech signal is processed by matched filtering,the interference noise is removed,and the English speech signal is processed by adaptive beam-focusing to enhance the intensity of the useful speech signal.The correlation spectrum detection method is used to extract the spectrum of the pronunciation speech signal of the speech spoken speech test system,and the speech pronunciation error detection is realized according to the spectrum difference.The simulation results show that the proposed method has the advantages of high accuracy and good anti-interference and error-correcting ability.

关 键 词：英语口语测试系统 语音信号 检测 发音错误 频谱 

分 类 号：TP391[自动化与计算机技术—计算机应用技术]