首页> 外国专利> METHOD AND SYSTEM FOR EVALUATING ENGLISH PRONUNCIATION

METHOD AND SYSTEM FOR EVALUATING ENGLISH PRONUNCIATION

机译:评估英语发音的方法和系统

摘要

The present invention relates to a method and a system for evaluating English pronunciation of a learner. The method and the system can automatically evaluate coincidence with a native speaker and the like in a phoneme and word level without using a voice recognition or a waveform comparison. The system of the present invention comprises: a voice evaluation engine having a predetermined native speaker database, analyzing a requested voice file, and providing an evaluation resu a plurality of learner terminals; a manager terminal; and an English phoneme evaluation server for setting an evaluation parameter in accordance with data transferred from the manager terminal via internet, displaying an evaluation sentence on the learner terminal via internet, assembling a voice file inputted by a learner into a predetermined request message, requesting evaluation to the voice evaluation engine, receiving and analyzing an evaluation result from the voice evaluation engine, and providing the evaluation result to a corresponding learner terminal suitably for the learner. The system according to the present invention does not use a sound recognition or a waveform comparison, but evaluates pronunciation of the learner on the basis of a large capacity of native speaker data in the continental U.S. and provides evaluation from various angles through accuracy, completeness, sentence confidence, native speaker coincidence (nativeness), the number of accurately pronounced words, the number of error words, a pronunciation speed, and the like, thereby being able to accurately evaluate pronunciation of the learner.;COPYRIGHT KIPO 2017
机译:用于评估学习者的英语发音的方法和系统技术领域本发明涉及一种用于评估学习者的英语发音的方法和系统。该方法和系统可以在不使用语音识别或波形比较的情况下自动评估与音素和单词级别中的母语者等的符合。本发明的系统包括:语音评估引擎,具有预定的母语者数据库,分析请求的语音文件,并提供评估结果;以及多个学习终端;经理终端;英文音素评估服务器,用于根据从管理者终端通过互联网传送的数据设置评估参数,通过互联网在学习者终端上显示评估语句,将学习者输入的语音文件组合成预定的请求消息,请求评估语音评估引擎接收评估结果,并从语音评估引擎分析评估结果,并将评估结果提供给适合学习者的相应学习者终端。根据本发明的系统不使用声音识别或波形比较,而是基于美国本土的大容量母语者数据来评估学习者的发音,并通过准确性,完整性,句子置信度,母语使用者的重合性(母语),准确发音的单词数量,错误单词的数量,发音速度等,从而能够准确地评估学习者的发音。; COPYRIGHT KIPO 2017

著录项

  • 公开/公告号KR20170056253A

    专利类型

  • 公开/公告日2017-05-23

    原文格式PDF

  • 申请/专利权人 LEE HO JIN;

    申请/专利号KR20150159693

  • 发明设计人 LEE HO JINKR;

    申请日2015-11-13

  • 分类号G09B19/06;G09B7/00;

  • 国家 KR

  • 入库时间 2022-08-21 13:27:27

相似文献

  • 专利
  • 外文文献
  • 中文文献
获取专利

客服邮箱:kefu@zhangqiaokeyan.com

京公网安备:11010802029741号 ICP备案号:京ICP备15016152号-6 六维联合信息科技 (北京) 有限公司©版权所有
  • 客服微信

  • 服务号