公开(公告)号：EP2541545A2

公开(公告)日：2013-01-02

申请号：EP12173408.1

申请日：2007-03-30

申请人： Vocollect, Inc.

发明人： Braho, Keith ,  Pike, Lori A. ,  Pike, Jeffrey

IPC分类号： G10L15/06

CPC分类号： G10L15/01 ,  G10L15/07

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.

摘要翻译： 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(公告)号：EP2541545A3

公开(公告)日：2013-09-04

申请号：EP12173408.1

申请日：2007-03-30

申请人： Vocollect, Inc.

发明人： Braho, Keith ,  Pike, Lori A. ,  Pike, Jeffrey

IPC分类号： G10L15/06

CPC分类号： G10L15/01 ,  G10L15/07

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.

公开(公告)号：EP2309490B1

公开(公告)日：2019-10-09

申请号：EP10179866.8

申请日：2006-02-02

申请人： Vocollect, Inc.

发明人： Pike, Jeffrey ,  El-Jaroudi, Amro ,  Braho, Keith

IPC分类号： G10L15/22 ,  G10L15/065

公开(公告)号：EP2309489B1

公开(公告)日：2018-11-21

申请号：EP10179864.3

申请日：2006-02-02

申请人： Vocollect, Inc.

发明人： Pike, Jeffrey ,  El-Jaroudi, Amro ,  Braho, Keith

IPC分类号： G10L15/22 ,  G10L15/06

CPC分类号： G10L15/22 ,  G10L15/065 ,  G10L2015/228

摘要： A speech recognition system receives and analyzes speech input from a user in order to recognize and accept a response from the user. Under certain conditions, information about the response expected from the user may be available, In these situations, the available information about the expected response is used to modify the behavior of the speech recognition system by taking this information into account. In one embodiment, the modified behavior of the speech recognition system comprises adjusting the rejection threshold when speech input matches the predetermined expected response. The modified behavior of the speech recognition system according to other embodiments of the invention include comparing the observed speech features to the models of the expected response separately from the usual hypothesis search in order to speed up the recognition system; modifying the usual hypothesis search to emphasize the expected response; updating and adapting the models when the recognized speech matches the expected response to improve the accuracy of the recognition system

公开(公告)号：EP2309490A1

公开(公告)日：2011-04-13

申请号：EP10179866.8

申请日：2006-02-02

申请人： Vocollect, Inc.

发明人： Pike, Jeffrey ,  El-Jaroudi, Amro ,  Braho, Keith

IPC分类号： G10L15/22 ,  G10L15/06

摘要： A speech recognition system receives and analyzes speech input from a user in order to recognize and accept a response from the user. Under certain conditions, information about the response expected from the user may be available, In these situations, the available information about the expected response is used to modify the behavior of the speech recognition system by taking this information into account. In one embodiment, the modified behavior of the speech recognition system comprises adjusting the rejection threshold when speech input matches the predetermined expected response. The modified behavior of the speech recognition system according to other embodiments of the invention include comparing the observed speech features to the models of the expected response separately from the usual hypothesis search in order to speed up the recognition system; modifying the usual hypothesis search to emphasize the expected response; updating and adapting the models when the recognized speech matches the expected response to improve the accuracy of the recognition system

摘要翻译： 一种语音识别系统接收并以识别和接受来自用户的响应来自用户的分析性的语音。 在某些条件下，关于从用户预期是可用的，在合成的情况下的响应的信息，关于所述预期响应的可用信息被用于通过取该信息考虑在内来修改所述语音识别系统的行为。 在一个，语音识别系统的实施例修改的行为包括：调整所述拒绝阈值。当版speechInput预定期望响应相匹配。 语音识别系统gemäß到本发明的其它实施例的修改的行为包括比较所观察到的语音特征来从以加快识别系统通常假设搜索分别预期响应的模型; 修改通常假设搜索以强调预期响应; 更新和适应的机型当识别的语音预期的响应相匹配，以提高识别系统的准确度

公开(公告)号：EP2309489A1

公开(公告)日：2011-04-13

申请号：EP10179864.3

申请日：2006-02-02

申请人： Vocollect, Inc.

发明人： Pike, Jeffrey ,  El-Jaroudi, Amro ,  Braho, Keith

IPC分类号： G10L15/22 ,  G10L15/06

CPC分类号： G10L15/22 ,  G10L15/065 ,  G10L2015/228

摘要： A speech recognition system receives and analyzes speech input from a user in order to recognize and accept a response from the user. Under certain conditions, information about the response expected from the user may be available. In these situations, the available information about the expected response is used to modify the behavior of the speech recognition system by taking this information into account. In one embodiment, the modified behavior of the speech recognition system comprises adjusting the rejection threshold when speech input matches the predetermined expected response. The modified behavior of the speech recognition system according to other embodiments of the invention includes comparing the observed speech features to the models of the expected response separately from the usual hypothesis search in order to speed up the recognition system; modifying the usual hypothesis search to emphasize the expected response; updating and adapting the models when the recognized speech matches the expected response to improve the accuracy of the recognition system

摘要翻译： 一种语音识别系统接收并以识别和接受来自用户的响应来自用户的分析性的语音。 在某些条件下，关于从用户预期响应信息可以是可用的。 在合成的情况下，对预期的反应可用的信息用于考虑到这一信息来修改语音识别系统的行为。 在一个，语音识别系统的实施例修改的行为包括：调整所述拒绝阈值。当版speechInput预定期望响应相匹配。 语音识别系统gemäß到本发明的其它实施例的修改的行为包括比较所观察到的语音特征来从以加快识别系统通常假设搜索分别预期响应的模型; 修改通常假设搜索以强调预期响应; 更新和适应的机型当识别的语音预期的响应相匹配，以提高识别系统的准确度