公开(公告)号：US08346549B2

公开(公告)日：2013-01-01

申请号：US12631131

申请日：2009-12-04

Applicant: Andrej Ljolje ,  Mazin Gilbert

Inventor： Andrej Ljolje ,  Mazin Gilbert

IPC: G10L15/00

CPC classification number: G10L15/00 ,  G10L15/285 ,  G10L15/32

Abstract: Disclosed herein are systems, methods, and computer-readable storage media for improving automatic speech recognition performance. A system practicing the method identifies idle speech recognition resources and establishes a supplemental speech recognizer on the idle resources based on overall speech recognition demand. The supplemental speech recognizer can differ from a main speech recognizer, and, along with the main speech recognizer, can be associated with a particular speaker. The system performs speech recognition on speech received from the particular speaker in parallel with the main speech recognizer and the supplemental speech recognizer and combines results from the main and supplemental speech recognizer. The system recognizes the received speech based on the combined results. The system can use beam adjustment in place of or in combination with a supplemental speech recognizer. A scheduling algorithm can tailor a particular combination of speech recognition resources and release the supplemental speech recognizer based on increased demand.

Abstract translation: 本文公开了用于改善自动语音识别性能的系统，方法和计算机可读存储介质。 实施该方法的系统识别空闲语音识别资源，并且基于总体语音识别需求在空闲资源上建立补充语音识别器。 补充语音识别器可以与主语音识别器不同，并且与主语音识别器一起可以与特定扬声器相关联。 该系统与主语音识别器和辅助语音识别器并行地执行从特定扬声器接收的语音的语音识别，并且组合来自主语音识别器和补充语音识别器的结果。 系统基于组合的结果识别接收到的语音。 该系统可以使用波束调整来代替或与补充语音识别器组合。 调度算法可以定制语音识别资源的特定组合，并且基于增加的需求来释放补充语音识别器。

公开(公告)号：US08195588B2

公开(公告)日：2012-06-05

申请号：US12080443

申请日：2008-04-03

Applicant: Narendra Gupta ,  Mazin Gilbert

Inventor： Narendra Gupta ,  Mazin Gilbert

IPC: G06Q20/00

CPC classification number: G06Q10/107

Abstract: Disclosed is a method and system for identifying critical emails. To identify critical emails, a critical email classifier is trained from training data comprising labeled emails. The classifier extracts N-grams from the training data and identifies N-gram features from the extracted N-grams. The classifier also extracts salient features from the training data. The classifier is trained based on the identified N-gram features and the salient features so that the classifier can classify unlabeled emails as critical emails or non-critical emails.

Abstract translation: 公开了用于识别关键电子邮件的方法和系统。 为了识别关键的电子邮件，关键的电子邮件分类器是从包含标记的电子邮件的数据的培训中进行培训的。 分类器从训练数据中提取N-gram，并从提取的N-gram中识别N-gram特征。 分类器还从训练数据中提取突出特征。 分类器基于识别的N-gram特征和突出特征进行训练，以便分类器可以将未标记的电子邮件分类为关键电子邮件或非关键电子邮件。

公开(公告)号：US20120101817A1

公开(公告)日：2012-04-26

申请号：US12908222

申请日：2010-10-20

Applicant: John MOCENIGO ,  Mazin Gilbert ,  Heather Wighton

Inventor： John MOCENIGO ,  Mazin Gilbert ,  Heather Wighton

IPC: G10L15/00

CPC classification number: G10L15/063 ,  G10L2015/0638

Abstract: Disclosed herein are systems, methods, and non-transitory computer-readable storage media for generating a model for use with automatic speech recognition. These principles can be implemented as part of a streamlined tool for automatic training and tuning of speech, or other, models with a fast turnaround and with limited human involvement. A system configured to practice the method receives, as part of a request to generate a model, input data and a seed model. The system receives a cost function indicating accuracy and at least one of speed and memory usage, The system processes the input data based on seed model and based on parameters that optimize the cost function to yield an updated model, and outputs the updated model.

Abstract translation: 这里公开了用于生成用于自动语音识别的模型的系统，方法和非暂时的计算机可读存储介质。 这些原则可以作为精简工具的一部分，用于自动训练和调整语音或其他模型，具有快速的周转时间和人力参与有限。 配置为练习该方法的系统作为生成模型的请求的一部分接收输入数据和种子模型。 系统接收指示精度和速度和存储器使用中的至少一个的成本函数。系统基于种子模型处理输入数据，并且基于优化成本函数以产生更新模型的参数，并输出更新的模型。

公开(公告)号：US20120065963A1

公开(公告)日：2012-03-15

申请号：US13300752

申请日：2011-11-21

Applicant: Srinivas Bangalore ,  Mazin Gilbert ,  Narendra Gupta

Inventor： Srinivas Bangalore ,  Mazin Gilbert ,  Narendra Gupta

IPC: G06F17/27

CPC classification number: G06F17/2785

Abstract: In accordance with one aspect of the present invention, an automated method of and system for generating a response to a text-based natural language message is disclosed. The method includes identifying a first selected input clause in a sentence in the text-based natural language message. Also, assigning a semantic tag to the first selected input clause and matching the semantic tag to a historical input tag. The historical input tag associated with a first previously generated response clause. Further; generating an output response message based on the historical response clause, the output response message derived from the historical input tag and a second previously generated response clause. The system includes means for performing the method steps.

Abstract translation: 根据本发明的一个方面，公开了一种用于生成对基于文本的自然语言消息的响应的自动化方法和系统。 该方法包括识别基于文本的自然语言消息中的句子中的第一选择的输入子句。 此外，将语义标签分配给第一选择的输入子句并将语义标签与历史输入标签进行匹配。 与先前生成的第一个响应子句相关联的历史输入标签。 进一步; 基于历史响应子句生成输出响应消息，从历史输入标签导出的输出响应消息和第二个先前生成的响应子句。 该系统包括用于执行方法步骤的装置。

公开(公告)号：US20120022950A1

公开(公告)日：2012-01-26

申请号：US12843836

申请日：2010-07-26

Applicant: Mazin Gilbert ,  Narendra K. Gupta ,  Dan Melamed

Inventor： Mazin Gilbert ,  Narendra K. Gupta ,  Dan Melamed

IPC: G06Q30/00 ,  G06F17/27 ,  H04M11/00

CPC classification number: H04M3/53333 ,  G06Q30/0241 ,  G06Q30/0269 ,  G06Q30/0273

Abstract: Systems and methods are provided for a voice message to text system supporting targeted advertisements. Voice messages received from users are converted to raw text messages that are normalized to insert proper punctuation and extract entity information. The normalized text and entity information are processed to extract concepts, such as critical phrases, from the normalized text. Extracted concepts are then matched to advertisements on an advertisement database having user selection criteria. Advertisements having selection criteria matching the extracted concepts are transmitted to the users, and the advertisers that placed the advertisements are charged fees for the advertisements. User profile information and user context information can additionally be used to select advertisements for transmission to users.

Abstract translation: 系统和方法被提供用于支持目标广告的文本系统的语音消息。 从用户接收的语音消息被转换为原始文本消息，其被标准化以插入适当的标点符号并提取实体信息。 处理归一化的文本和实体信息以从标准化文本中提取概念，例如关键短语。 然后将提取的概念与具有用户选择标准的广告数据库上的广告相匹配。 将具有与所提取的概念相匹配的选择标准的广告传送给用户，并且放置广告的广告商对广告收取费用。 用户简档信息和用户上下文信息可另外用于选择用于传输给用户的广告。

公开(公告)号：US20110321098A1

公开(公告)日：2011-12-29

申请号：US12823734

申请日：2010-06-25

Applicant: Srinivas Bangalore ,  Mazin Gilbert ,  Michael Johnston

Inventor： Srinivas Bangalore ,  Mazin Gilbert ,  Michael Johnston

IPC: H04N5/445 ,  H04N7/173

CPC classification number: H04N21/44222 ,  G06F17/30743 ,  G06F17/30787 ,  G06F17/30864 ,  G06F17/30964 ,  H04N7/0884 ,  H04N21/2355 ,  H04N21/435 ,  H04N21/4394 ,  H04N21/44008 ,  H04N21/440236 ,  H04N21/4532 ,  H04N21/4622 ,  H04N21/47815 ,  H04N21/4828 ,  H04N21/4884 ,  H04N21/8405

Abstract: An Internet Protocol television system includes a user profile agent, a keyword detection agent, and an information search agent. The user profile agent is in communication with a multimedia device, and generates a user profile based on information received from the multimedia device. The keyword detection agent is in communication with the user profile agent, and searches text associated with a multimedia video stream transmitted to the multimedia device for keywords associated with the user profile. The information search agent is in communication with the keyword detection agent, and connects to an information source associated with the keywords detected by the keyword detection agent, and provides additional information associated with the keywords to the multimedia device.

Abstract translation: 互联网协议电视系统包括用户简档代理，关键字检测代理和信息搜索代理。 用户简档代理与多媒体设备通信，并且基于从多媒体设备接收的信息来生成用户简档。 关键字检测代理与用户简档代理进行通信，并且搜索与传输到多媒体设备的多媒体视频流相关联的文本，以获得与用户简档相关联的关键字。 信息搜索代理与关键字检测代理进行通信，并且连接到与由关键字检测代理检测到的关键字相关联的信息源，并向多媒体设备提供与关键字相关联的附加信息。

公开(公告)号：US20110184973A1

公开(公告)日：2011-07-28

申请号：US13007931

申请日：2011-01-17

Applicant: SRINIVAS BANGALORE ,  Junlan Feng ,  Mazin Gilbert

Inventor： SRINIVAS BANGALORE ,  Junlan Feng ,  Mazin Gilbert

IPC: G06F17/30

CPC classification number: G06F17/3089

Abstract: A method and apparatus for automatically detecting and extracting information from dynamically generated web pages are disclosed. For example, the present method stores user provided information that is entered into a form interlace of a web page for a first query. Responsive to the first query, a first response web page is received and stored. The present method then automatically generates a second query to acquire a second response web page that is responsive to the second query. Finally, the present method compares the first response web page and the second response web page. In one embodiment, the present invention extracts information that is dissimilar between the first response web page and the second response web page. This extracted information is deemed to be the pertinent information requested by the user.

Abstract translation: 公开了一种用于从动态生成的网页自动检测和提取信息的方法和装置。 例如，本方法将用户提供的信息存储在用于第一查询的网页的交错格式中。 响应于第一个查询，收到并存储第一个响应网页。 本方法然后自动生成第二查询以获取响应于第二查询的第二响应网页。 最后，本方法比较第一响应网页和第二响应网页。 在一个实施例中，本发明提取在第一响应网页和第二响应网页之间不相似的信息。 该提取的信息被认为是用户请求的相关信息。

公开(公告)号：US20090254498A1

公开(公告)日：2009-10-08

申请号：US12080443

申请日：2008-04-03

Applicant: Narendra Gupta ,  Mazin Gilbert

Inventor： Narendra Gupta ,  Mazin Gilbert

IPC: G06F15/18

CPC classification number: G06Q10/107

Abstract: Disclosed is a method and system for identifying critical emails. To identify critical emails, a critical email classifier is trained from training data comprising labeled emails. The classifier extracts N-grams from the training data and identifies N-gram features from the extracted N-grams. The classifier also extracts salient features from the training data. The classifier is trained based on the identified N-gram features and the salient features so that the classifier can classify unlabeled emails as critical emails or non-critical emails.

Abstract translation: 公开了用于识别关键电子邮件的方法和系统。 为了识别关键的电子邮件，关键的电子邮件分类器是从包含标记的电子邮件的数据的培训中进行培训的。 分类器从训练数据中提取N-gram，并从提取的N-gram中识别N-gram特征。 分类器还从训练数据中提取突出特征。 分类器基于识别的N-gram特征和突出特征进行训练，以便分类器可以将未标记的电子邮件分类为关键电子邮件或非关键电子邮件。

公开(公告)号：US20090112600A1

公开(公告)日：2009-04-30

申请号：US11931830

申请日：2007-10-31

Applicant: Mazin Gilbert ,  Christopher Volinsky

Inventor： Mazin Gilbert ,  Christopher Volinsky

IPC: G10L11/00

CPC classification number: G10L15/193 ,  G10L15/065 ,  G10L2015/228

Abstract: Disclosed are systems, methods and computer-readable media for using a local communication network to generate a speech model. The method includes retrieving for an individual a list of numbers in a calling history, identifying a local neighborhood associated with each number in the calling history, truncating the local neighborhood associated with each number based on the at least one parameter, retrieving a local communication network associated with each number in the calling history and each phone number in the local neighborhood, and creating a language model for the individual based on the retrieved local communication network. The generated language model may be used for improved automatic speech recognition for audible searches as well as other modules in a spoken dialog system.

Abstract translation: 公开了用于使用本地通信网络生成语音模型的系统，方法和计算机可读介质。 该方法包括检索个人在呼叫历史中的号码列表，识别与呼叫历史中的每个号码相关联的本地邻域，基于该至少一个参数截断与每个号码相关联的本地邻域，检索本地通信网络 与呼叫历史中的每个号码和本地邻域中的每个电话号码相关联，以及基于所检索的本地通信网络为个人创建语言模型。 生成的语言模型可以用于改善声音搜索的自动语音识别以及口语对话系统中的其他模块。

公开(公告)号：US20090076795A1

公开(公告)日：2009-03-19

申请号：US11857036

申请日：2007-09-18

Applicant: Srinivas Bangalore ,  Mazin Gilbert ,  Narendra Gupta

Inventor： Srinivas Bangalore ,  Mazin Gilbert ,  Narendra Gupta

IPC: G06F17/27

CPC classification number: G06F17/2785

Abstract: In accordance with one aspect of the present invention, an automated method of and system for generating a response to a text-based natural language message is disclosed. The method includes identifying a sentence in the text-based natural language message. Also, identifying an input clause in the sentence. Further, comparing the input clause to a previously received clause, where the previously received clause is correlated with a previously generated response message. Additionally, generating an output response message based on the previously generated response message. The system includes means for performing the method steps.

Abstract translation: 根据本发明的一个方面，公开了一种用于生成对基于文本的自然语言消息的响应的自动化方法和系统。 该方法包括识别基于文本的自然语言消息中的句子。 另外，确定句子中的一个input子句。 此外，将输入子句与先前接收的子句进行比较，其中先前接收的子句与先前生成的响应消息相关联。 另外，基于先前生成的响应消息生成输出响应消息。 该系统包括用于执行方法步骤的装置。