首页> 外国专利> Method for speaker identification for telecommunication network, involves ascertaining data on the identity of a person characterized by a search profile

Method for speaker identification for telecommunication network, involves ascertaining data on the identity of a person characterized by a search profile

机译:用于电信网络的说话者识别的方法,包括确定有关以搜索配置文件为特征的人的身份的数据

摘要

A method for speaker identification uses an identification system based on verbal information which is linked to a knowledge-based system in which by means of a search routine, preferably via a search-engine from a publicly accessible data-bank (7) and/or at least one internal customer data bank on the basis of person-related search profile installed in the search routine, data on the identity of the person characterized by the search profile is ascertained and from the data ascertained for the person, suitable data is extracted via a filter function for the speaker recognition enquiry, which is defined as an answer for the speaker identification process. Data suitable as a answer for the identification process is automatically stored for speaker verification together with the answers. An independent claim is included for a system for knowledge-based speaker identification.
机译:用于说话者识别的方法使用基于语言信息的识别系统,该识别系统链接到基于知识的系统,在该系统中,借助于搜索例程,优选地经由搜索引擎,来自公众可访问的数据库(7)和/或基于安装在搜索例程中的与人相关的搜索配置文件,至少一个内部客户数据库,确定以该搜索配置文件为特征的人的身份数据,并从为该人确定的数据中提取合适的数据,用于说话人识别查询的过滤器功能,它被定义为说话人识别过程的答案。自动将适合作为识别过程答案的数据与答案一起存储以用于说话者验证。包括针对基于知识的说话者识别的系统的独立主张。

著录项

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