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

US08812321B2 System and method for combining speech recognition outputs from a plurality of domain-specific speech recognizers via machine learning 有权
标题翻译：用于通过机器学习来组合来自多个领域特定语音识别器的语音识别输出的系统和方法
公开(公告)号：US08812321B2
公开(公告)日：2014-08-19
申请号：US12895359
申请日：2010-09-30
申请人： Mazin Gilbert , Srinivas Bangalore , Patrick Haffner , Robert Bell
发明人： Mazin Gilbert , Srinivas Bangalore , Patrick Haffner , Robert Bell
IPC分类号： G10L15/08 , G10L15/32
CPC分类号： G10L15/32 , G10L15/063 , G10L15/26 , G10L2015/0638
摘要： Disclosed herein are systems, methods and non-transitory computer-readable media for performing speech recognition across different applications or environments without model customization or prior knowledge of the domain of the received speech. The disclosure includes recognizing received speech with a collection of domain-specific speech recognizers, determining a speech recognition confidence for each of the speech recognition outputs, selecting speech recognition candidates based on a respective speech recognition confidence for each speech recognition output, and combining selected speech recognition candidates to generate text based on the combination.
摘要翻译：本文公开了用于在不需要模型定制或接收到的语音的领域的先前知识的情况下在不同的应用或环境上执行语音识别的系统，方法和非暂时的计算机可读介质。该公开内容包括：利用特定领域的语音识别器的集合来识别接收的语音，为每个语音识别输出确定语音识别置信度，基于每个语音识别输出的相应的语音识别置信度选择语音识别候选，以及组合所选语音识别候选人基于组合生成文本。

2. 发明申请

US20120253799A1 SYSTEM AND METHOD FOR RAPID CUSTOMIZATION OF SPEECH RECOGNITION MODELS 有权
标题翻译：用于快速识别语音识别模型的系统和方法
公开(公告)号：US20120253799A1
公开(公告)日：2012-10-04
申请号：US13072920
申请日：2011-03-28
申请人： Srinivas BANGALORE , Robert Bell , Diamantino Antonio Caseiro , Mazin Gilbert , Patrick Haffner
发明人： Srinivas BANGALORE , Robert Bell , Diamantino Antonio Caseiro , Mazin Gilbert , Patrick Haffner
IPC分类号： G10L15/00
CPC分类号： G10L15/183 , G10L15/06 , G10L15/065 , G10L15/22 , G10L15/30 , G10L2015/0635 , G10L2015/0636 , G10L2015/228
摘要： Disclosed herein are systems, methods, and non-transitory computer-readable storage media for generating domain-specific speech recognition models for a domain of interest by combining and tuning existing speech recognition models when a speech recognizer does not have access to a speech recognition model for that domain of interest and when available domain-specific data is below a minimum desired threshold to create a new domain-specific speech recognition model. A system configured to practice the method identifies a speech recognition domain and combines a set of speech recognition models, each speech recognition model of the set of speech recognition models being from a respective speech recognition domain. The system receives an amount of data specific to the speech recognition domain, wherein the amount of data is less than a minimum threshold to create a new domain-specific model, and tunes the combined speech recognition model for the speech recognition domain based on the data.
摘要翻译：本文公开的是当语音识别器不能访问语音识别模型时通过组合和调整现有语音识别模型来生成针对感兴趣域的域特定语音识别模型的系统，方法和非暂时计算机可读存储介质对于该感兴趣的领域，并且当可用的域特定数据低于创建新的特定领域语音识别模型的最小期望阈值时。配置为练习该方法的系统识别语音识别域并组合一组语音识别模型，语音识别模型集合的每个语音识别模型来自相应的语音识别域。系统接收特定于语音识别域的数据量，其中数据量小于最小阈值以创建新的域特定模型，并且基于该数据调谐用于语音识别域的组合语音识别模型。

3. 发明申请

US20120084086A1 SYSTEM AND METHOD FOR OPEN SPEECH RECOGNITION 有权
标题翻译：用于开放语音识别的系统和方法
公开(公告)号：US20120084086A1
公开(公告)日：2012-04-05
申请号：US12895359
申请日：2010-09-30
申请人： Mazin GILBERT , Srinivas Bangalore , Patrick Haffner , Robert Bell
发明人： Mazin GILBERT , Srinivas Bangalore , Patrick Haffner , Robert Bell
IPC分类号： G10L15/26
CPC分类号： G10L15/32 , G10L15/063 , G10L15/26 , G10L2015/0638
摘要： Disclosed herein are systems, methods and non-transitory computer-readable media for performing speech recognition across different applications or environments without model customization or prior knowledge of the domain of the received speech. The disclosure includes recognizing received speech with a collection of domain-specific speech recognizers, determining a speech recognition confidence for each of the speech recognition outputs, selecting speech recognition candidates based on a respective speech recognition confidence for each speech recognition output, and combining selected speech recognition candidates to generate text based on the combination.
摘要翻译：本文公开了用于在不需要模型定制或接收到的语音的领域的先前知识的情况下在不同的应用或环境上执行语音识别的系统，方法和非暂时的计算机可读介质。该公开内容包括：利用特定领域的语音识别器的集合来识别接收的语音，为每个语音识别输出确定语音识别置信度，基于每个语音识别输出的相应语音识别置信度选择语音识别候选，以及组合所选语音识别候选人基于组合生成文本。

4. 发明授权

US09679561B2 System and method for rapid customization of speech recognition models 有权
公开(公告)号：US09679561B2
公开(公告)日：2017-06-13
申请号：US13072920
申请日：2011-03-28
申请人： Srinivas Bangalore , Robert Bell , Diamantino Antonio Caseiro , Mazin Gilbert , Patrick Haffner
发明人： Srinivas Bangalore , Robert Bell , Diamantino Antonio Caseiro , Mazin Gilbert , Patrick Haffner
IPC分类号： G10L15/183 , G10L15/30 , G10L15/22 , G10L15/06 , G10L15/065
CPC分类号： G10L15/183 , G10L15/06 , G10L15/065 , G10L15/22 , G10L15/30 , G10L2015/0635 , G10L2015/0636 , G10L2015/228
摘要： Disclosed herein are systems, methods, and non-transitory computer-readable storage media for generating domain-specific speech recognition models for a domain of interest by combining and tuning existing speech recognition models when a speech recognizer does not have access to a speech recognition model for that domain of interest and when available domain-specific data is below a minimum desired threshold to create a new domain-specific speech recognition model. A system configured to practice the method identifies a speech recognition domain and combines a set of speech recognition models, each speech recognition model of the set of speech recognition models being from a respective speech recognition domain. The system receives an amount of data specific to the speech recognition domain, wherein the amount of data is less than a minimum threshold to create a new domain-specific model, and tunes the combined speech recognition model for the speech recognition domain based on the data.

5. 发明申请

US20070244688A1 On-Demand Language Translation For Television Programs 有权
标题翻译：电视节目的按需语言翻译
公开(公告)号：US20070244688A1
公开(公告)日：2007-10-18
申请号：US11279852
申请日：2006-04-14
申请人： Srinivas Bangalore , David Gibbon , Mazin Gilbert , Patrick Haffner , Zhu Liu , Behzad Shahraray
发明人： Srinivas Bangalore , David Gibbon , Mazin Gilbert , Patrick Haffner , Zhu Liu , Behzad Shahraray
IPC分类号： G06F17/28
CPC分类号： H04N21/4856 , G06F17/289 , G10L21/10 , H04N7/0885 , H04N7/17336 , H04N21/2355 , H04N21/2543 , H04N21/25435 , H04N21/4355 , H04N21/440236 , H04N21/4882 , H04N21/4884 , H04N21/8193
摘要： A method, a system and a machine-readable medium are provided for an on demand translation service. A translation module including at least one language pair module for translating a source language to a target language may be made available for use by a subscriber. The subscriber may be charged a fee for use of the requested on demand translation service or may be provided use of the on demand translation service for free in exchange for displaying commercial messages to the subscriber. A video signal may be received including information in the source language, which may be obtained as text from the video signal and may be translated from the source language to the target language by use of the translation module. Translated information, based on the translated text, may be added into the received video signal. The video signal including the translated information in the target language may be sent to a display device.
摘要翻译：为按需翻译服务提供方法，系统和机器可读介质。包括用于将源语言翻译成目标语言的至少一个语言对模块的翻译模块可以被用户使用。用户可能会收取使用所请求的按需翻译服务的费用，或者可以免费使用按需翻译服务，以便向用户显示商业消息。可以接收包括源语言的信息的视频信号，其可以从视频信号获取为文本，并且可以通过使用翻译模块从源语言转换为目标语言。基于翻译文本的翻译信息可以被添加到接收的视频信号中。可以将包含目标语言的翻译信息的视频信号发送到显示装置。

6. 发明申请

US20080162117A1 Discriminative training of models for sequence classification 审中-公开
标题翻译：序列分类模型的辨别性训练
公开(公告)号：US20080162117A1
公开(公告)日：2008-07-03
申请号：US11646983
申请日：2006-12-28
申请人： Srinivas Bangalore , Patrick Haffner , Stephan Kanthak
发明人： Srinivas Bangalore , Patrick Haffner , Stephan Kanthak
IPC分类号： G06F17/21
CPC分类号： G06F17/2818
摘要： Classification of sequences, such as the translation of natural language sentences, is carried out using an independence assumption. The independence assumption is an assumption that the probability of a correct translation of a source sentence word into a particular target sentence word is independent of the translation of other words in the sentence. Although this assumption is not a correct one, a high level of word translation accuracy is nonetheless achieved. In particular, discriminative training is used to develop models for each target vocabulary word based on a set of features of the corresponding source word in training sentences, with at least one of those features relating to the context of the source word. Each model comprises a weight vector for the corresponding target vocabulary word. The weights comprising the vectors are associated with respective ones of the features; each weight is a measure of the extent to which the presence of that feature for the source word makes it more probable that the target word in question is the correct one.
摘要翻译：使用独立假设进行序列分类，如自然语言句子的翻译。独立性假设是将源语句正确翻译成特定目标句子词的概率与句子中其他单词的翻译无关的假设。尽管这种假设不是正确的，但仍然会实现高水平的字翻译精度。特别地，歧视性训练被用于基于训练句子中相应源词的一组特征来开发每个目标词汇词的模型，其中至少一个与源词的上下文有关的特征。每个模型包括对应的目标词汇单词的权重向量。包括向量的权重与相应的特征相关联; 每个权重是衡量源字符的该特征的存在程度使得所述目标词更可能是正确的。

7. 发明申请

US20100169244A1 METHOD AND APPARATUS FOR USING A DISCRIMINATIVE CLASSIFIER FOR PROCESSING A QUERY 有权
标题翻译：使用分辨率分类器处理查询的方法和装置
公开(公告)号：US20100169244A1
公开(公告)日：2010-07-01
申请号：US12347513
申请日：2008-12-31
申请人： Ilija Zeljkovic , Srinivas Bangalore , Patrick Haffner , Jay Wilpon
发明人： Ilija Zeljkovic , Srinivas Bangalore , Patrick Haffner , Jay Wilpon
IPC分类号： G06F15/18 , G06F17/30
CPC分类号： G06F17/30867 , G06F17/30011 , G06F17/30648 , G06F17/3071 , G06F17/30864 , G06N99/005
摘要： A method and apparatus for using a classifier for processing a query are disclosed. For example, the method receives a query from a user, and processes the query to locate one or more documents in accordance with a search engine having a discriminative classifier, wherein the discriminative classifier is trained with a plurality of artificial query examples. The method then presents a result of the processing to the user.
摘要翻译：公开了一种使用分类器处理查询的方法和装置。例如，该方法从用户接收查询，并且根据具有辨别性分类器的搜索引擎处理该查询以定位一个或多个文档，其中用多个人工查询示例训练该辨别分类器。该方法然后向用户呈现处理的结果。

8. 发明申请

US20080162111A1 Sequence classification for machine translation 失效
标题翻译：机器翻译序列分类
公开(公告)号：US20080162111A1
公开(公告)日：2008-07-03
申请号：US11647080
申请日：2006-12-28
申请人： Srinivas Bangalore , Patrick Haffner , Stephan Kanthak
发明人： Srinivas Bangalore , Patrick Haffner , Stephan Kanthak
IPC分类号： G06F17/28
CPC分类号： G06F17/2818
摘要： Classification of sequences, such as the translation of natural language sentences, is carried out using an independence assumption. The independence assumption is an assumption that the probability of a correct translation of a source sentence word into a particular target sentence word is independent of the translation of other words in the sentence. Although this assumption is not a correct one, a high level of word translation accuracy is nonetheless achieved. In particular, discriminative training is used to develop models for each target vocabulary word based on a set of features of the corresponding source word in training sentences, with at least one of those features relating to the context of the source word. Each model comprises a weight vector for the corresponding target vocabulary word. The weights comprising the vectors are associated with respective ones of the features; each weight is a measure of the extent to which the presence of that feature for the source word makes it more probable that the target word in question is the correct one.
摘要翻译：使用独立假设进行序列分类，如自然语言句子的翻译。独立性假设是将源语句正确翻译成特定目标句子词的概率与句子中其他单词的翻译无关的假设。尽管这种假设不是正确的，但仍然会实现高水平的字翻译精度。特别地，歧视性训练被用于基于训练句子中相应源词的一组特征来开发每个目标词汇词的模型，其中至少一个与源词的上下文有关的特征。每个模型包括对应的目标词汇单词的权重向量。包括向量的权重与相应的特征相关联; 每个权重是衡量源字符的该特征的存在程度使得所述目标词更可能是正确的。

9. 发明授权

US08799279B2 Method and apparatus for using a discriminative classifier for processing a query 有权
标题翻译：用于使用辨别性分类器处理查询的方法和装置
公开(公告)号：US08799279B2
公开(公告)日：2014-08-05
申请号：US12347513
申请日：2008-12-31
申请人： Ilija Zeljkovic , Srinivas Bangalore , Patrick Haffner , Jay Wilpon
发明人： Ilija Zeljkovic , Srinivas Bangalore , Patrick Haffner , Jay Wilpon
IPC分类号： G06F7/00 , G06F17/30
CPC分类号： G06F17/30867 , G06F17/30011 , G06F17/30648 , G06F17/3071 , G06F17/30864 , G06N99/005
摘要： A method and apparatus for using a classifier for processing a query are disclosed. For example, the method receives a query from a user, and processes the query to locate one or more documents in accordance with a search engine having a discriminative classifier, wherein the discriminative classifier is trained with a plurality of artificial query examples. The method then presents a result of the processing to the user.
摘要翻译：公开了一种使用分类器处理查询的方法和装置。例如，该方法从用户接收查询，并且根据具有辨别性分类器的搜索引擎处理该查询以定位一个或多个文档，其中用多个人工查询示例训练该辨别分类器。该方法然后向用户呈现处理的结果。

10. 发明授权

US08788258B1 Machine translation using global lexical selection and sentence reconstruction 有权
标题翻译：机器翻译使用全局词法选择和句子重建
公开(公告)号：US08788258B1
公开(公告)日：2014-07-22
申请号：US11686681
申请日：2007-03-15
申请人： Srinivas Bangalore , Patrick Haffner , Stephan Kanthak
发明人： Srinivas Bangalore , Patrick Haffner , Stephan Kanthak
IPC分类号： G06F17/28
CPC分类号： G06F17/2818 , G06F17/2827 , G06F17/289 , G10L15/26 , G11B27/105 , G11B27/28 , H04N5/44591 , H04N5/45 , H04N5/765 , H04N21/232 , H04N21/233 , H04N21/235 , H04N21/25825 , H04N21/2662 , H04N21/4828 , H04N21/812 , H04N21/84 , H04N21/8456 , H04N21/8547
摘要： Disclosed are systems, methods, and computer-readable media for performing translations from a source language to a target language. The method comprises receiving a source phrase, generating a target bag of words based on a global lexical selection of words that loosely couples the source words/phrases and target words/phrases, and reconstructing a target phrase or sentence by considering all permutations of words with a conditional probability greater than a threshold.
摘要翻译：公开了用于执行从源语言到目标语言的翻译的系统，方法和计算机可读介质。该方法包括接收源短语，基于松散地耦合源词/短语和目标词/短语的单词的全局词汇选择来产生目标包词，以及通过考虑所有词的排列来重建目标短语或句子，条件概率大于阈值。

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