首页> 外国专利> SYSTEM HAVING COMPOSITE STATISTICS MODEL AND RULE-BASED GRAMMAR MODEL FOR SPEECH RECOGNITION AND NATURAL LANGUAGE UNDERSTANDING

SYSTEM HAVING COMPOSITE STATISTICS MODEL AND RULE-BASED GRAMMAR MODEL FOR SPEECH RECOGNITION AND NATURAL LANGUAGE UNDERSTANDING

机译:具有用于语音识别和自然语言理解的综合统计模型和基于规则的语法模型的系统

摘要

PURPOSE: A system having a composite statistics and rule based grammar model is provided to perform both of speech recognition operation and natural language understanding using a composite statistics model and a rule-based grammar language model. CONSTITUTION: A model authoring system(200) includes a model authoring component(202) and an optional user interface(204). The model authoring component receives schema(206), a training illustration text string(208), and an optional grammar library(209) as inputs and outputs a rule-based grammar(210). The optional grammar library includes a definition for an area-dependent concept obtained from an application database and an area-independent concept. A user provides the schema and the training illustration text string to the model authoring component through the optional user interface.
机译:目的:提供一种具有基于复合统计和基于规则的语法模型的系统,以使用复合统计模型和基于规则的语法语言模型来执行语音识别操作和自然语言理解。构成:模型创作系统(200)包括模型创作组件(202)和可选的用户界面(204)。模型创作组件接收模式(206),训练插图文本字符串(208)和可选的语法库(209)作为输入,并输出基于规则的语法(210)。可选的语法库包括从应用程序数据库获得的区域相关概念的定义和区域独立概念。用户通过可选的用户界面将模式和训练插图文本字符串提供给模型创作组件。

著录项

  • 公开/公告号KR20040094642A

    专利类型

  • 公开/公告日2004-11-10

    原文格式PDF

  • 申请/专利权人 MICROSOFT CORP.;

    申请/专利号KR20040030526

  • 发明设计人 WANG YE YI;ACERO ALEJANDRO;CHELBA CIPRIAN;

    申请日2004-04-30

  • 分类号G10L15/18;

  • 国家 KR

  • 入库时间 2022-08-21 22:47:40

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