首页> 外国专利> Dynamic grammatical weighting method for speech recognition system has existing probability distribution for grammatical entries modified for each identified user

Dynamic grammatical weighting method for speech recognition system has existing probability distribution for grammatical entries modified for each identified user

机译:用于语音识别系统的动态语法加权方法具有针对每个识别出的用户修改的语法条目的现有概率分布

摘要

The grammatical weighting method identifies a user of the speech recognition system and provides an individual probability distribution for grammatic entries by adapting an existing probability distribution via a user profile for the identified user, the individual probability distribution employed by the speech recognition system during a spoken input. Also included are Independent claims for the following: (a) a computer program for a grammatical weighting method for a speech recognition system; (b) a speech identification system; (c) an application of a speech recognition system with grammatical weighting for automatic banking transactions
机译:语法加权方法可以识别语音识别系统的用户,并通过针对已识别用户的用户配置文件调整现有的概率分布,从而为语法条目提供单独的概率分布,语音输入系统在语音输入过程中采用的单独概率分布。还包括以下方面的独立权利要求:(a)用于语音识别系统的语法加权方法的计算机程序; (b)语音识别系统; (c)具有语法加权的语音识别系统在自动银行交易中的应用

著录项

  • 公开/公告号DE10131157C1

    专利类型

  • 公开/公告日2002-07-04

    原文格式PDF

  • 申请/专利权人 PROJECT49 AG;

    申请/专利号DE2001131157

  • 发明设计人 UNTERKIRCHER FLORIAN M.;

    申请日2001-06-29

  • 分类号G10L15/00;

  • 国家 DE

  • 入库时间 2022-08-22 00:26:57

相似文献

  • 专利
  • 外文文献
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