首页> 外文会议>World Multi-conference on Systemics, Cybernetics and Informatics >Key-Phrase Spotting Technique Used in Telephone Dialog System Accessing E-mails via Voice
【24h】

Key-Phrase Spotting Technique Used in Telephone Dialog System Accessing E-mails via Voice

机译:通过语音访问电子邮件的电话对话系统中使用的关键类型

获取原文

摘要

This paper discusses a usage of key-phrase spotting technique implemented in a real-time telephone dialog system designed and constructed at the Department of Cybernetics, University of West Bohemia, and its contribution to speech recognition and understanding. The paper is concerned about e-mail retrieving and reading service, which is controlled through the telephone. Our approach is to give to the user more convenience and allow him/her to spoke with the system. Users can control the system by predefined phrases. The problem arising in this case is that the user is not able to use only one predefined phrase at the given dialog state but rather tell the system his/her request nearly in a naturally way. To permit user to communicate more naturally and to control the system more intuitively we implemented a simple understanding module based on a key-phrase spotting technique. The paper describes an implementation of the key-phrase spotting method using a mumble model, a finite state grammar format and the structure of the mumble model. The paper also shows how special dialog applications can handle language understanding using the key-phrase spotting technique and VoiceXML. Results achieved during the experimental system usage and testing conditions are included.
机译:本文讨论了在网络对象系统,西波希米大学的实时电话对话系统中实施的关键类型的专题技术的用法,以及对语音识别和理解的贡献。本文涉及通过电话控制的电子邮件检索和读取服务。我们的方法是向用户提供更多便利,让他/她与系统交谈。用户可以通过预定义的短语来控制系统。在这种情况下出现的问题是用户无法在给定的对话状态下仅在一个预定义的短语中使用,而是略先以自然的方式告诉系统他/她的请求。允许用户更自然地沟通并更直观地控制系统,我们基于密钥短语点发现技术实现了一个简单的理解模块。本文描述了使用炫目模型的关键短语点发现方法的实现,有限状态语法格式和炫目模型的结构。本文还展示了特殊的对话框应用程序如何使用密钥短语点发现技术和VoiceXML处理语言理解。包括在实验系统使用和测试条件下实现的结果。

著录项

相似文献

  • 外文文献
  • 中文文献
  • 专利
获取原文

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