专利快速检索-快速检索全球专利，免费商用专利数据库-IPRDB

1. 发明授权

US08306822B2 Automatic reading tutoring using dynamically built language model 有权
标题翻译：使用动态内置语言模型进行自动阅读辅导
公开(公告)号：US08306822B2
公开(公告)日：2012-11-06
申请号：US11900274
申请日：2007-09-11
申请人： Xiaolong Li , Li Deng , Yun-Cheng Ju , Alex Acero
发明人： Xiaolong Li , Li Deng , Yun-Cheng Ju , Alex Acero
IPC分类号： G10L15/22
CPC分类号： G10L15/18 , G09B17/003 , G10L15/183
摘要： A method of providing automatic reading tutoring is disclosed. The method includes retrieving a textual indication of a story from a data store and creating a language model including constructing a target context free grammar indicative of a first portion of the story. A first acoustic input is received and a speech recognition engine is employed to recognize the first acoustic input. An output of the speech recognition engine is compared to the language model and a signal indicative of whether the output of the speech recognition matches at least a portion of the target context free grammar is provided.
摘要翻译：公开了提供自动阅读辅导的方法。该方法包括从数据存储中检索故事的文本指示并创建包括构建指示故事的第一部分的目标上下文自由语法的语言模型。接收到第一声输入，并且采用语音识别引擎来识别第一声输入。将语音识别引擎的输出与语言模型进行比较，并且提供指示语音识别的输出是否与目标语境自由语法的至少一部分匹配的信号。

2. 发明授权

US07813926B2 Training system for a speech recognition application 有权
标题翻译：用于语音识别应用的训练系统
公开(公告)号：US07813926B2
公开(公告)日：2010-10-12
申请号：US11376853
申请日：2006-03-16
申请人： Ye-Yi Wang , John Sie Yuen Lee , Alex Acero
发明人： Ye-Yi Wang , John Sie Yuen Lee , Alex Acero
IPC分类号： G10L15/00 , G10L15/06 , G10L15/18
CPC分类号： G10L15/063
摘要： A training system for a speech recognition application is disclosed. In embodiments described, the training system is used to train a classification model or language model. The classification model is trained using an adaptive language model generated by an iterative training process. In embodiments described, the training data is recognized by the speech recognition component and the recognized text is used to create the adaptive language model which is used for speech recognition in a following training iteration.
摘要翻译：公开了一种用于语音识别应用的训练系统。在所描述的实施例中，训练系统用于训练分类模型或语言模型。使用由迭代训练过程产生的自适应语言模型训练分类模型。在所描述的实施例中，训练数据由语音识别组件识别，并且识别的文本用于创建在随后的训练迭代中用于语音识别的自适应语言模型。

3. 发明申请

US20080052075A1 Incrementally regulated discriminative margins in MCE training for speech recognition 有权
标题翻译：增加对语音识别的MCE训练中的歧视性空白
公开(公告)号：US20080052075A1
公开(公告)日：2008-02-28
申请号：US11509980
申请日：2006-08-25
申请人： Xiaodong He , Alex Acero , Dong Yu , Li Deng
发明人： Xiaodong He , Alex Acero , Dong Yu , Li Deng
IPC分类号： G10L15/14
CPC分类号： G10L15/063 , G10L15/144
摘要： A method and apparatus for training an acoustic model are disclosed. A training corpus is accessed and converted into an initial acoustic model. Scores are calculated for a correct class and competitive classes, respectively, for each token given the acoustic model. From this score a misclassification measure is calculated and then a loss function is calculated from the misclassification measure. The loss function also includes a margin value that varies over each iteration in the training. Based on the calculated loss function the acoustic model is updated, where the loss function with the margin value is minimized. This process repeats until such time as an empirical convergence is met.
摘要翻译：公开了一种用于训练声学模型的方法和装置。训练语料库被访问并转换成初始声学模型。对于给定声学模型的每个令牌，分数是针对正确的班级和竞赛班分别计算的。从该分数计算错误分类度量，然后根据误分类度量计算损失函数。损失函数还包括在训练中每次迭代变化的保证金值。基于计算的损耗函数，声学模型被更新，其中具有边际值的损失函数被最小化。该过程重复，直到满足经验收敛的时间为止。

4. 发明申请

US20080004880A1 PERSONALIZED SPEECH SERVICES ACROSS A NETWORK 审中-公开
标题翻译：通过网络进行个性化的语音服务
公开(公告)号：US20080004880A1
公开(公告)日：2008-01-03
申请号：US11424459
申请日：2006-06-15
申请人： Alex Acero , Timothy S. Paek , Christopher A. Meek , David M. Chickering
发明人： Alex Acero , Timothy S. Paek , Christopher A. Meek , David M. Chickering
IPC分类号： G10L21/00
CPC分类号： G10L15/22 , G10L2015/227
摘要： A speech application accessible across a network is personalized for a particular user based on preferences for the user. The speech application can be modified based on the preferences.
摘要翻译：基于对用户的偏好，可以针对特定用户个性化跨网络访问的语音应用。语音应用可以根据喜好进行修改。

5. 发明申请

US20070219798A1 Training system for a speech recognition application 有权
标题翻译：用于语音识别应用的训练系统
公开(公告)号：US20070219798A1
公开(公告)日：2007-09-20
申请号：US11376853
申请日：2006-03-16
申请人： Ye-Yi Wang , John Lee , Alex Acero
发明人： Ye-Yi Wang , John Lee , Alex Acero
IPC分类号： G10L15/18
CPC分类号： G10L15/063
摘要： A training system for a speech recognition application is disclosed. In embodiments described, the training system is used to train a classification model or language model. The classification model is trained using an adaptive language model generated by an iterative training process. In embodiments described, the training data is recognized by the speech recognition component and the recognized text is used to create the adaptive language model which is used for speech recognition in a following training iteration.
摘要翻译：公开了一种用于语音识别应用的训练系统。在所描述的实施例中，训练系统用于训练分类模型或语言模型。使用由迭代训练过程产生的自适应语言模型训练分类模型。在所描述的实施例中，训练数据由语音识别组件识别，并且识别的文本用于创建在随后的训练迭代中用于语音识别的自适应语言模型。

6. 发明申请

US20090070112A1 Automatic reading tutoring 有权
标题翻译：自动阅读辅导
公开(公告)号：US20090070112A1
公开(公告)日：2009-03-12
申请号：US11900274
申请日：2007-09-11
申请人： Xiaolong Li , Li Deng , Yun-Cheng Ju , Alex Acero
发明人： Xiaolong Li , Li Deng , Yun-Cheng Ju , Alex Acero
IPC分类号： G10L15/18
CPC分类号： G10L15/18 , G09B17/003 , G10L15/183
摘要： A method of providing automatic reading tutoring is disclosed. The method includes retrieving a textual indication of a story from a data store and creating a language model including constructing a target context free grammar indicative of a first portion of the story. A first acoustic input is received and a speech recognition engine is employed to recognize the first acoustic input. An output of the speech recognition engine is compared to the language model and a signal indicative of whether the output of the speech recognition matches at least a portion of the target context free grammar is provided.
摘要翻译：公开了提供自动阅读辅导的方法。该方法包括从数据存储中检索故事的文本指示并创建包括构建指示故事的第一部分的目标上下文自由语法的语言模型。接收到第一声输入，并且采用语音识别引擎来识别第一声输入。将语音识别引擎的输出与语言模型进行比较，并且提供指示语音识别的输出是否与目标语境自由语法的至少一部分匹配的信号。

7. 发明申请

US20090043497A1 Conveying Locations In Spoken Dialog Systems 有权
标题翻译：输入口语对话系统中的位置
公开(公告)号：US20090043497A1
公开(公告)日：2009-02-12
申请号：US11836955
申请日：2007-08-10
申请人： Ivan Tashev , Michael Lewis Seltzer , Yun-Cheng Ju , Alex Acero
发明人： Ivan Tashev , Michael Lewis Seltzer , Yun-Cheng Ju , Alex Acero
IPC分类号： G01C21/34
CPC分类号： G01C21/3644 , G01C21/3679
摘要： The presentation of location information to a user that is distracted by traveling can result in the user quickly forgetting, or never even comprehending, key parts of the location information, such as the street number. Identification can be made of intersections and points of interest near the user's destination, which can then be provided instead of, or in addition to, the address, thereby increasing user comprehension and retention, especially when distracted. Map data can be parsed into addresses, intersections and points of interest databases. These databases can be accessed to identify proximate intersections and points of interest, which can then be filtered and subsequently ranked to identify one intersection, one point of interest, or both, that can be presented to the user to aid the user in comprehending and retaining the location information even when distracted.
摘要翻译：通过旅行分散给用户的位置信息的呈现可能导致用户快速地忘记甚至不理解诸如街道号码的位置信息的关键部分。识别可以由用户目的地附近的交叉点和兴趣点组成，然后可以提供地址，也可以除了地址之外，还可以提供用户的理解和保留，特别是在分心时。地图数据可以解析为地址，交叉点和兴趣点数据库。可以访问这些数据库以识别最近的交叉点和兴趣点，然后可以对这些数据进行过滤并随后进行排序以识别一个交点，一个兴趣点或二者，可以呈现给用户以帮助用户理解和保留位置信息即使分心。

8. 发明授权

US08065078B2 Conveying locations in spoken dialog systems 有权
标题翻译：在口语对话系统中传送位置
公开(公告)号：US08065078B2
公开(公告)日：2011-11-22
申请号：US11836955
申请日：2007-08-10
申请人： Ivan Tashev , Michael Lewis Seltzer , Yun-Cheng Ju , Alex Acero
发明人： Ivan Tashev , Michael Lewis Seltzer , Yun-Cheng Ju , Alex Acero
IPC分类号： G01C21/00 , G08G1/123
CPC分类号： G01C21/3644 , G01C21/3679
摘要： The presentation of location information to a user that is distracted by traveling can result in the user quickly forgetting, or never even comprehending, key parts of the location information, such as the street number. Identification can be made of intersections and points of interest near the user's destination, which can then be provided instead of, or in addition to, the address, thereby increasing user comprehension and retention, especially when distracted. Map data can be parsed into addresses, intersections and points of interest databases. These databases can be accessed to identify proximate intersections and points of interest, which can then be filtered and subsequently ranked to identify one intersection, one point of interest, or both, that can be presented to the user to aid the user in comprehending and retaining the location information even when distracted.
摘要翻译：通过旅行分散给用户的位置信息的呈现可能导致用户快速地忘记甚至不理解诸如街道号码的位置信息的关键部分。识别可以由用户目的地附近的交叉点和兴趣点组成，然后可以提供地址，也可以除了地址之外提供，从而增加用户理解和保留，特别是在分心时。地图数据可以解析为地址，交叉点和兴趣点数据库。可以访问这些数据库以识别最近的交叉点和兴趣点，然后可以对这些数据进行过滤并随后进行排序以识别一个交点，一个兴趣点或二者，可以呈现给用户以帮助用户理解和保留位置信息即使分心。

9. 发明授权

US07617103B2 Incrementally regulated discriminative margins in MCE training for speech recognition 有权
标题翻译：增加对语音识别的MCE训练中的歧视性空白
公开(公告)号：US07617103B2
公开(公告)日：2009-11-10
申请号：US11509980
申请日：2006-08-25
申请人： Xiaodong He , Alex Acero , Dong Yu , Li Deng
发明人： Xiaodong He , Alex Acero , Dong Yu , Li Deng
IPC分类号： G10L15/14
CPC分类号： G10L15/063 , G10L15/144
摘要： A method and apparatus for training an acoustic model are disclosed. A training corpus is accessed and converted into an initial acoustic model. Scores are calculated for a correct class and competitive classes, respectively, for each token given the acoustic model. From this score a misclassification measure is calculated and then a loss function is calculated from the misclassification measure. The loss function also includes a margin value that varies over each iteration in the training. Based on the calculated loss function the acoustic model is updated, where the loss function with the margin value is minimized. This process repeats until such time as an empirical convergence is met.
摘要翻译：公开了一种用于训练声学模型的方法和装置。训练语料库被访问并转换成初始声学模型。对于给定声学模型的每个令牌，分数是针对正确的班级和竞赛班分别计算的。从该分数计算错误分类度量，然后根据误分类度量计算损失函数。损失函数还包括在训练中每次迭代变化的保证金值。基于计算的损耗函数，声学模型被更新，其中具有边际值的损失函数被最小化。该过程重复，直到满足经验收敛的时间为止。

10. 发明申请

US20090248422A1 INTRA-LANGUAGE STATISTICAL MACHINE TRANSLATION 有权
标题翻译：语言统计机翻译
公开(公告)号：US20090248422A1
公开(公告)日：2009-10-01
申请号：US12058328
申请日：2008-03-28
申请人： Xiao Li , Yun-Cheng Ju , Geoffrey Zweig , Alex Acero
发明人： Xiao Li , Yun-Cheng Ju , Geoffrey Zweig , Alex Acero
IPC分类号： G10L11/00 , G06F17/28
CPC分类号： G06F17/2818 , G06F17/2827
摘要： Training data may be provided, the training data including pairs of source phrases and target phrases. The pairs may be used to train an intra-language statistical machine translation model, where the intra-language statistical machine translation model, when given an input phrase of text in the human language, can compute probabilities of semantic equivalence of the input phrase to possible translations of the input phrase in the human language. The statistical machine translation model may be used to translate between queries and listings. The queries may be text strings in the human language submitted to a search engine. The listing strings may be text strings of formal names of real world entities that are to be searched by the search engine to find matches for the query strings.
摘要翻译：可以提供训练数据，训练数据包括源短语和目标短语对。这些对可以用于训练语言间统计机器翻译模型，其中语言内统计机器翻译模型在给予人类语言的文本的输入短语时可以计算输入短语的语义等同性的可能性输入短语在人类语言中的翻译。统计机器翻译模型可用于在查询和列表之间进行翻译。查询可以是提交给搜索引擎的人类语言中的文本字符串。列表字符串可以是要由搜索引擎搜索以查找查询字符串的匹配的真实世界实体的正式名称的文本串。

你已经成功收藏专利！

检索式保存成功!

IPRDB

热门服务

关于我们

友情链接

联系方式