基于单目标跟踪算法的发音自动校准系统设计  

Design of pronunciation automatic calibration system based on single target tracking algorithm

作　　者：李玉华[1] LI Yuhua(College of Science and Technology,Nanchang University,Nanchang 330029,China)

机构地区：[1]南昌大学科学技术学院,江西南昌330029

出　　处：《现代电子技术》2018年第13期151-154,共4页Modern Electronics Technique

基　　金：江西省级教改课题:网络环境下独立学院非英语专业大学生英语听力学习策略的研究与实践(JXJG-08-78-17);江西省社科课题J.K罗琳笔下"魔法学校"的魅力--哈利波特系列小说教育理念研究(11WX328)~~

摘　　要：传统发音校准系统存在英语发音校准准确率低的问题,采用单目标跟踪算法设计发音自动校准系统,利用UNIX风格子程序对自动校准系统硬件框架进行构建,遵循内高聚合原则分析数据资源提取模块流程。针对英语发音信息采集需利用模拟数字信号转换来提高数据采样效率,对发音校准引擎A/D电路进行设计。采用单目标跟踪算法提取相关特征,并形成逻辑层。开发嵌入式内核结构,研究语音识别代码。通过实验验证可知,该系统英语发音校准准确率高,对发音纠正能力较强。The traditional pronunciation calibration system has low accuracy for English pronunciation calibration.Therefore,the single target tracking algorithm is adopted to design the pronunciation automatic calibration system. The UNIXstyle subroutine is used to build the hardware framework of the automatic calibration system,which follows the high-polymeric principle to analyze the extraction process of the data resource extraction module. The A/D signal conversion is used for English pronunciation information acquisition to improve the data sampling efficiency,and its circuit is designed. The single target tracking algorithm is adopted to extract the related features to form a logical layer. The embedded kernel structure is developed,and the speech recognition code is studied. The experimental results show that the system has high accuracy for English pronunciation calibration,and strong pronunciation correction ability.

关 键 词：单目标跟踪算法 发音 自动校准 A/D电路 逻辑层 语音识别 

分 类 号：TN912.34[电子电信—通信与信息系统]