利索-全球专利检索

  1. 登录
  2. 注册
发明公开
EP2541545A2 Methods and systems for adapting a model for a speech recognition system 审中-公开
Verfahren und Systeme zur adaptationment eines Modellsfürein Spracherkennungssystem

  • 专利标题： Methods and systems for adapting a model for a speech recognition system
  • 专利标题（中）： Verfahren und Systeme zur adaptationment eines Modellsfürein Spracherkennungssystem
  • 申请号： EP12173408.1
    申请日： 2007-03-30
  • 公开(公告)号： EP2541545A2
    公开(公告)日： 2013-01-02
  • 发明人: Braho, Keith , Pike, Lori A. , Pike, Jeffrey
  • 申请人： Vocollect, Inc.
  • 申请人地址： 703 Rodi Road Pittsburgh, Pennsylvania 15235 US
  • 专利权人： Vocollect, Inc.
  • 当前专利权人： Vocollect, Inc.
  • 当前专利权人地址： 703 Rodi Road Pittsburgh, Pennsylvania 15235 US
  • 代理机构： Findlay, Alice Rosemary
  • 优先权： US788621P 20060403; US788622P 20060403; US788606P 20060403; US539456 20061006; US688920 20070321; US688916 20070321
  • 主分类号： G10L15/06
  • IPC分类号： G10L15/06
Methods and systems for adapting a model for a speech recognition system
摘要：
Methods are disclosed for identifying possible errors made by a speech recognition system without using a transcript of words input to the system. A method for model adaptation for a speech recognition system includes determining an error rate (315), corresponding to either recognition of instances of a word or recognition of instances of various words, without using a transcript of words input to the system. The method may further include adjusting an adaptation 320, of the model for the word or various models for the various words, based on the error rate. Apparatus are disclosed for identifying possible errors made by a speech recognition system without using a transcript of words input to the system. An apparatus for model adaptation for a speech recognition system includes a processor 210 adapted to estimate an error rate, corresponding to either recognition of instances of a word or recognition of instances of various words, without using a transcript of words input to the system. The apparatus may further include a controller 225 adapted to adjust an adaptation of the model 235 for the word or various models for the various words, based on the error rate.
摘要（中）：

公开了用于识别由语音识别系统做出的可能的错误的方法，而不使用输入到系统的单词的抄本。 一种用于语音识别系统的模型适应的方法包括：确定误差率（315），对应于识别单词的实例或识别各种单词的实例，而不使用输入到系统的单词的抄本。 该方法可以进一步包括基于错误率来调整用于单词的模型的模型或适用于各种单词的各种模型的自适应320。 公开了用于识别由语音识别系统进行的可能的错误的装置，而不使用输入到系统的单词的抄本。 用于语音识别系统的用于模型适配的装置包括处理器210，其适于估计对应于识别单词的实例或识别各种单词的实例的错误率，而不使用输入到系统的单词的抄本。 该装置可以进一步包括控制器225，该控制器225适于基于错误率来调整针对各种单词的单词或各种模型的模型235的适配。

公开/授权文献
信息查询
Global Dossier Espacenet
0/0