公开(公告)号：US09263031B2

公开(公告)日：2016-02-16

申请号：US14081166

申请日：2013-11-15

Applicant: AT&T Intellectual Property II, L.P.

Inventor： Srinivas Bangalore ,  Narendra K. Gupta ,  Mazin G. Rahim

IPC: G10L21/00 ,  G06F17/20 ,  G06F17/27 ,  G10L15/02

CPC classification number: G10L15/1815 ,  G06F17/2785 ,  G10L13/043 ,  G10L15/02 ,  G10L15/1822 ,  G10L15/265

Abstract: A system and method are disclosed that improve automatic speech recognition in a spoken dialog system. The method comprises partitioning speech recognizer output into self-contained clauses, identifying a dialog act in each of the self-contained clauses, qualifying dialog acts by identifying a current domain object and/or a current domain action, and determining whether further qualification is possible for the current domain object and/or current domain action. If further qualification is possible, then the method comprises identifying another domain action and/or another domain object associated with the current domain object and/or current domain action, reassigning the another domain action and/or another domain object as the current domain action and/or current domain object and then recursively qualifying the new current domain action and/or current object. This process continues until nothing is left to qualify.

公开(公告)号：US20130311170A1

公开(公告)日：2013-11-21

申请号：US13873548

申请日：2013-04-30

Applicant: AT&T Intellectual Property II, L.P.

Inventor： Srinivas Bangalore ,  Mazin Gilbert ,  Narendra K. Gupta

IPC: G06F17/28

CPC classification number: G10L15/183 ,  G06F17/2818 ,  G10L15/14 ,  G10L15/19

Abstract: Disclosed herein are systems and methods to incorporate human knowledge when developing and using statistical models for natural language understanding. The disclosed systems and methods embrace a data-driven approach to natural language understanding which progresses seamlessly along the continuum of availability of annotated collected data, from when there is no available annotated collected data to when there is any amount of annotated collected data.

Abstract translation: 这里公开的是在开发和使用自然语言理解的统计模型时将人类知识结合在一起的系统和方法。 所公开的系统和方法包含一种数据驱动的自然语言理解方法，其从注释收集的数据的可用性的连续性无缝地进行，从没有可用的注释收集的数据到当有任何数量的注释收集的数据时。

公开(公告)号：US20130166284A1

公开(公告)日：2013-06-27

申请号：US13775546

申请日：2013-02-25

Applicant: AT&T Intellectual Property II, L.P.

Inventor： Srinivas Bangalore ,  Narendra K. Gupta ,  Mazin G. Rahim

IPC: G06F17/27

CPC classification number: G10L15/1815 ,  G06F17/2785 ,  G10L13/043 ,  G10L15/02 ,  G10L15/1822 ,  G10L15/265

Abstract: A system and method are disclosed that improve automatic speech recognition in a spoken dialog system. The method comprises partitioning speech recognizer output into self-contained clauses, identifying a dialog act in each of the self-contained clauses, qualifying dialog acts by identifying a current domain object and/or a current domain action, and determining whether further qualification is possible for the current domain object and/or current domain action. If further qualification is possible, then the method comprises identifying another domain action and/or another domain object associated with the current domain object and/or current domain action, reassigning the another domain action and/or another domain object as the current domain action and/or current domain object and then recursively qualifying the new current domain action and/or current object. This process continues until nothing is left to qualify.

Abstract translation: 公开了一种提高口语对话系统中的自动语音识别的系统和方法。 该方法包括将语音识别器输出划分为独立子句，识别每个自包含子句中的对话行为，通过识别当前域对象和/或当前域动作进行限定对话行为，以及确定是否可进一步限定 对于当前域对象和/或当前域操作。 如果可以进一步鉴定，则该方法包括识别与当前域对象和/或当前域操作相关联的另一域操作和/或另一域对象，将另一域操作和/或另一域对象重新分配为当前域操作，以及 /或当前域对象，然后递归地限定新的当前域操作和/或当前对象。 这个过程一直持续到没有什么是剩下的资格。

公开(公告)号：US09548051B2

公开(公告)日：2017-01-17

申请号：US15014070

申请日：2016-02-03

Applicant: AT&T Intellectual Property II, L.P.

Inventor： Srinivas Bangalore ,  Narendra K. Gupta ,  Mazin G. Rahim

IPC: G10L21/00 ,  G06F17/20 ,  G06F17/27 ,  G10L15/18 ,  G10L15/02 ,  G10L13/04 ,  G10L15/26

CPC classification number: G10L15/1815 ,  G06F17/2785 ,  G10L13/043 ,  G10L15/02 ,  G10L15/1822 ,  G10L15/265

Abstract: A system and method are disclosed that improve automatic speech recognition in a spoken dialog system. The method comprises partitioning speech recognizer output into self-contained clauses, identifying a dialog act in each of the self-contained clauses, qualifying dialog acts by identifying a current domain object and/or a current domain action, and determining whether further qualification is possible for the current domain object and/or current domain action. If further qualification is possible, then the method comprises identifying another domain action and/or another domain object associated with the current domain object and/or current domain action, reassigning the another domain action and/or another domain object as the current domain action and/or current domain object and then recursively qualifying the new current domain action and/or current object. This process continues until nothing is left to qualify.

Abstract translation: 公开了一种提高口语对话系统中的自动语音识别的系统和方法。 该方法包括将语音识别器输出划分为独立子句，识别每个自包含子句中的对话行为，通过识别当前域对象和/或当前域动作进行限定对话行为，以及确定是否可进行进一步的限定 对于当前域对象和/或当前域操作。 如果可以进一步鉴定，则该方法包括识别与当前域对象和/或当前域操作相关联的另一域操作和/或另一域对象，将另一域操作和/或另一域对象重新分配为当前域动作，以及 /或当前域对象，然后递归地限定新的当前域操作和/或当前对象。 这个过程一直持续到没有什么是剩下的资格。

公开(公告)号：US09215305B2

公开(公告)日：2015-12-15

申请号：US14532213

申请日：2014-11-04

Applicant: AT&T Intellectual Property II, L.P.

Inventor： Narendra K. Gupta ,  Benjamin J. Stern

IPC: H04M1/64 ,  H04M1/665 ,  H04M1/27 ,  H04M1/652 ,  H04M3/533 ,  H04M1/60 ,  H04Q1/45

CPC classification number: H04M1/7255 ,  G10L17/005 ,  G10L17/24 ,  H04M1/271 ,  H04M1/6041 ,  H04M1/652 ,  H04M1/665 ,  H04M3/53333 ,  H04M2201/40 ,  H04Q1/45

Abstract: An arrangement for allowing “hands-free” access to DTMF-controlled systems, such as one's voice mail messaging systems, utilizes a speech-to-DTMF tone application that monitors the communication between the user and the DTMF-controlled system. A speech recognition unit is utilized to retrieve certain voice commands (e.g., “next”, “skip”, “repeat”, “forward”, etc.) when uttered by the user. The application then translates the received commands into the proper DTMF tone sequence used by the DTMF-controlled system and transmits the DTMF tones to the system. The application is particularly useful in the cell phone environment and avoids the necessity of the user to constantly switch between using the keypad and listening to messages/commands from the system.

Abstract translation: 允许“免提”访问DTMF控制的系统（例如语音邮件消息系统）的安排利用语音到DTMF音调应用程序来监视用户和DTMF控制的系统之间的通信。 语音识别单元用于在用户发出时检索某些语音命令（例如，“下一个”，“跳过”，“重复”，“前进”等）。 然后，应用程序将接收到的命令转换为由DTMF控制的系统使用的正确的DTMF音调序列，并将DTMF音调发送到系统。 该应用在蜂窝电话环境中特别有用，并且避免了用户在使用键盘和从系统收听消息/命令之间不断地切换的必要性。

公开(公告)号：US09176946B2

公开(公告)日：2015-11-03

申请号：US14499888

申请日：2014-09-29

Applicant: AT&T Intellectual Property II, L.P.

Inventor： Srinivas Bangalore ,  Narendra K. Gupta ,  Mazin G. Rahim

IPC: G06F17/27 ,  G06F17/21 ,  G10L15/26

CPC classification number: G06F17/2705 ,  G06F17/218 ,  G06F17/27 ,  G06F17/2775 ,  G10L15/05 ,  G10L15/26 ,  G10L2015/025 ,  G10L2015/081 ,  G10L2015/088

Abstract: A clausifier and method of extracting clauses for spoken language understanding are disclosed. The method relates to generating a set of clauses from speech utterance text and comprises inserting at least one boundary tag in speech utterance text related to sentence boundaries, inserting at least one edit tag indicating a portion of the speech utterance text to remove, and inserting at least one conjunction tag within the speech utterance text. The result is a set of clauses that may be identified within the speech utterance text according to the inserted at least one boundary tag, at least one edit tag and at least one conjunction tag. The disclosed clausifier comprises a sentence boundary classifier, an edit detector classifier, and a conjunction detector classifier. The clausifier may comprise a single classifier or a plurality of classifiers to perform the steps of identifying sentence boundaries, editing text, and identifying conjunctions within the text.

Abstract translation: 公开了一种提取语言理解条款的分类器和方法。 该方法涉及从语音话语文本生成一组子句，并且包括在与句子边界相关的语音话语文本中插入至少一个边界标签，插入指示语音话语文本的一部分的至少一个编辑标签以移除，并插入到 讲话话语文本内的至少一个连接标签。 结果是可以根据插入的至少一个边界标签，至少一个编辑标签和至少一个连接标签在语音发音文本内识别的一组子句。 所公开的分类器包括句子边界分类器，编辑检测器分类器和连接检测器分类器。 克隆器可以包括单个分类器或多个分类器，以执行识别句子边界，编辑文本以及识别文本内的连词的步骤。

公开(公告)号：US20140029746A1

公开(公告)日：2014-01-30

申请号：US13959228

申请日：2013-08-05

Applicant: AT&T Intellectual Property II, L.P.

Inventor： Giuseppe Di Fabbrizio ,  Narendra K. Gupta

IPC: H04M1/2745

CPC classification number: H04M1/27455 ,  H04M7/003 ,  H04M15/06 ,  H04M15/83 ,  H04M15/8351 ,  H04M15/8353 ,  H04M15/84 ,  H04M15/844 ,  H04M15/846 ,  H04M2215/81 ,  H04M2215/8108 ,  H04M2215/8112 ,  H04M2215/8129 ,  H04M2215/8137 ,  H04M2215/8141

Abstract: A system and method to provide content and call attributes for a destination phone number using a click-to-dial connection includes accepting a query, retrieving links to a document, and searching through cached data using the query to identify relevant or destination number information. If no match is found, the method continues with accessing the document identified by the link for identifying relevant number information and creating a click-to-dial icon to link to the relevant numbers included in the relevant phone number information identified by the query. Next, a popup box for content relevant to the click-to-dial icon and a click feature on the popup box are created to retrieve call destination attributes for viewing by a user. Further, computer instructions create at least one communication connection between two communication devices after viewing call destination attributes.

Abstract translation: 使用点击拨号连接为目的地电话号码提供内容和呼叫属性的系统和方法包括接受查询，检索到文档的链接，以及使用该查询来搜索缓存的数据以识别相关或目的地号码信息。 如果没有找到匹配，则该方法继续访问用于识别相关号码信息的链接识别的文档，并创建点击拨号图标以链接到由查询识别的相关电话号码信息中包括的相关号码。 接下来，创建用于与点击拨号图标相关的内容和弹出框上的点击功能的弹出框，以检索呼叫目的地属性以供用户查看。 此外，计算机指令在查看呼叫目的地属性之后，在两个通信设备之间创建至少一个通信连接。

公开(公告)号：US20140011544A1

公开(公告)日：2014-01-09

申请号：US14021452

申请日：2013-09-09

Applicant: AT&T Intellectual Property II, L.P.

Inventor： Narendra K. Gupta ,  Benjamin J. Stern

IPC: H04M1/60

CPC classification number: H04M1/7255 ,  G10L17/005 ,  G10L17/24 ,  H04M1/271 ,  H04M1/6041 ,  H04M1/652 ,  H04M1/665 ,  H04M3/53333 ,  H04M2201/40 ,  H04Q1/45

Abstract: An arrangement for allowing “hands-free” access to DTMF-controlled systems, such as one's voice mail messaging systems, utilizes a speech-to-DTMF tone application that monitors the communication between the user and the DTMF-controlled system. A speech recognition unit is utilized to retrieve certain voice commands (e.g., “next”, “skip”, “repeat”, “forward”, etc.) when uttered by the user. The application then translates the received commands into the proper DTMF tone sequence used by the DTMF-controlled system and transmits the DTMF tones to the system. The application is particularly useful in the cell phone environment and avoids the necessity of the user to constantly switch between using the keypad and listening to messages/commands from the system.

Abstract translation: 允许“免提”访问DTMF控制的系统（例如语音邮件消息系统）的安排利用语音到DTMF音调应用程序来监视用户和DTMF控制的系统之间的通信。 语音识别单元用于在用户发出时检索某些语音命令（例如，“下一个”，“跳过”，“重复”，“前进”等）。 然后，应用程序将接收到的命令转换为由DTMF控制的系统使用的正确的DTMF音调序列，并将DTMF音调发送到系统。 该应用在蜂窝电话环境中特别有用，并且避免了用户在使用键盘和从系统收听消息/命令之间不断地切换的必要性。

公开(公告)号：US08612232B2

公开(公告)日：2013-12-17

申请号：US13775546

申请日：2013-02-25

Applicant: AT&T Intellectual Property II, L.P.

Inventor： Srinivas Bangalore ,  Narendra K. Gupta ,  Mazin G. Rahim

IPC: G10L21/00 ,  G06F17/20 ,  G06F17/27

CPC classification number: G10L15/1815 ,  G06F17/2785 ,  G10L13/043 ,  G10L15/02 ,  G10L15/1822 ,  G10L15/265

Abstract: A system and method are disclosed that improve automatic speech recognition in a spoken dialog system. The method comprises partitioning speech recognizer output into self-contained clauses, identifying a dialog act in each of the self-contained clauses, qualifying dialog acts by identifying a current domain object and/or a current domain action, and determining whether further qualification is possible for the current domain object and/or current domain action. If further qualification is possible, then the method comprises identifying another domain action and/or another domain object associated with the current domain object and/or current domain action, reassigning the another domain action and/or another domain object as the current domain action and/or current domain object and then recursively qualifying the new current domain action and/or current object. This process continues until nothing is left to qualify.

公开(公告)号：US08934619B2

公开(公告)日：2015-01-13

申请号：US13959228

申请日：2013-08-05

Applicant: AT&T Intellectual Property II, L.P.

Inventor： Giuseppe Di Fabbrizio ,  Narendra K. Gupta

IPC: H04M1/00 ,  H04M1/2745 ,  H04M7/00 ,  H04M15/06 ,  H04M15/00

CPC classification number: H04M1/27455 ,  H04M7/003 ,  H04M15/06 ,  H04M15/83 ,  H04M15/8351 ,  H04M15/8353 ,  H04M15/84 ,  H04M15/844 ,  H04M15/846 ,  H04M2215/81 ,  H04M2215/8108 ,  H04M2215/8112 ,  H04M2215/8129 ,  H04M2215/8137 ,  H04M2215/8141

Abstract: A system and method to provide content and call attributes for a destination phone number using a click-to-dial connection includes accepting a query, retrieving links to a document, and searching through cached data using the query to identify relevant or destination number information. If no match is found, the method continues with accessing the document identified by the link for identifying relevant number information and creating a click-to-dial icon to link to the relevant numbers included in the relevant phone number information identified by the query. Next, a popup box for content relevant to the click-to-dial icon and a click feature on the popup box are created to retrieve call destination attributes for viewing by a user. Further, computer instructions create at least one communication connection between two communication devices after viewing call destination attributes.

Abstract translation: 使用点击拨号连接为目的地电话号码提供内容和呼叫属性的系统和方法包括接受查询，检索到文档的链接，以及使用该查询来搜索缓存的数据以识别相关或目的地号码信息。 如果没有找到匹配，则该方法继续访问用于识别相关号码信息的链接识别的文档，并创建点击拨号图标以链接到由查询识别的相关电话号码信息中包括的相关号码。 接下来，创建用于与点击拨号图标相关的内容和弹出框上的点击功能的弹出框，以检索呼叫目的地属性以供用户查看。 此外，计算机指令在查看呼叫目的地属性之后，在两个通信设备之间创建至少一个通信连接。