-
公开(公告)号：US08566090B2
公开(公告)日：2013-10-22
申请号：US13465685
申请日：2012-05-07
IPC分类号： G10L13/027
CPC分类号： G06F17/2881 , G06F17/279 , G10L13/027
摘要： Systems, methods, and non-transitory computer-readable media for referring to entities. The method includes receiving domain-specific training data of sentences describing a target entity in a context, extracting a speaker history and a visual context from the training data, selecting attributes of the target entity based on at least one of the speaker history, the visual context, and speaker preferences, generating a text expression referring to the target entity based on at least one of the selected attributes, the speaker history, and the context, and outputting the generated text expression. The weighted finite-state automaton can represent partial orderings of word pairs in the domain-specific training data. The weighted finite-state automaton can be speaker specific or speaker independent. The weighted finite-state automaton can include a set of weighted partial orderings of the training data for each possible realization.
摘要翻译： 用于引用实体的系统，方法和非暂时计算机可读介质。 该方法包括接收在上下文中描述目标实体的句子的特定领域的训练数据，从训练数据中提取讲者历史和视觉上下文，基于说话者的历史，视觉上的至少一个来选择目标实体的属性 上下文和说话人首选项，基于所选择的属性，说话者历史和上下文中的至少一个生成参考目标实体的文本表达，并输出所生成的文本表达。 加权有限状态自动机可以表示域特定训练数据中单词对的部分排序。 加权有限状态自动机可以是扬声器专用或扬声器独立的。 加权有限状态自动机可以包括用于每个可能实现的训练数据的一组加权部分排序。
-
公开(公告)号：US08175873B2
公开(公告)日：2012-05-08
申请号：US12333863
申请日：2008-12-12
CPC分类号： G06F17/2881 , G06F17/279 , G10L13/027
摘要： Systems, methods, and non-transitory computer-readable media for referring to entities. The method includes receiving domain-specific training data of sentences describing a target entity in a context, extracting a speaker history and a visual context from the training data, selecting attributes of the target entity based on at least one of the speaker history, the visual context, and speaker preferences, generating a text expression referring to the target entity based on at least one of the selected attributes, the speaker history, and the context, and outputting the generated text expression. The weighted finite-state automaton can represent partial orderings of word pairs in the domain-specific training data. The weighted finite-state automaton can be speaker specific or speaker independent. The weighted finite-state automaton can include a set of weighted partial orderings of the training data for each possible realization.
摘要翻译： 用于引用实体的系统，方法和非暂时计算机可读介质。 该方法包括接收在上下文中描述目标实体的句子的特定领域的训练数据，从训练数据中提取讲者历史和视觉上下文，基于说话者的历史，视觉上的至少一个来选择目标实体的属性 上下文和说话人首选项，基于所选择的属性，说话者历史和上下文中的至少一个生成参考目标实体的文本表达，并输出所生成的文本表达。 加权有限状态自动机可以表示域特定训练数据中单词对的部分排序。 加权有限状态自动机可以是扬声器专用或扬声器独立的。 加权有限状态自动机可以包括用于每个可能实现的训练数据的一组加权部分排序。
-
公开(公告)号：US20120221332A1
公开(公告)日：2012-08-30
申请号：US13465685
申请日：2012-05-07
IPC分类号： G10L15/26
CPC分类号： G06F17/2881 , G06F17/279 , G10L13/027
摘要： Systems, methods, and non-transitory computer-readable media for referring to entities. The method includes receiving domain-specific training data of sentences describing a target entity in a context, extracting a speaker history and a visual context from the training data, selecting attributes of the target entity based on at least one of the speaker history, the visual context, and speaker preferences, generating a text expression referring to the target entity based on at least one of the selected attributes, the speaker history, and the context, and outputting the generated text expression. The weighted finite-state automaton can represent partial orderings of word pairs in the domain-specific training data. The weighted finite-state automaton can be speaker specific or speaker independent. The weighted finite-state automaton can include a set of weighted partial orderings of the training data for each possible realization.
摘要翻译： 用于引用实体的系统，方法和非暂时计算机可读介质。 该方法包括接收在上下文中描述目标实体的句子的特定领域的训练数据，从训练数据中提取讲者历史和视觉上下文，基于说话者的历史，视觉上的至少一个来选择目标实体的属性 上下文和说话人首选项，基于所选择的属性，说话者历史和上下文中的至少一个生成参考目标实体的文本表达，并输出所生成的文本表达。 加权有限状态自动机可以表示域特定训练数据中单词对的部分排序。 加权有限状态自动机可以是扬声器专用或扬声器独立的。 加权有限状态自动机可以包括用于每个可能实现的训练数据的一组加权部分排序。
-
公开(公告)号：US20100153105A1
公开(公告)日：2010-06-17
申请号：US12333863
申请日：2008-12-12
IPC分类号： G10L15/26
CPC分类号： G06F17/2881 , G06F17/279 , G10L13/027
摘要： Disclosed herein are systems, computer-implemented methods, and tangible computer-readable media for referring to entities. The method includes receiving domain-specific training data of sentences describing a target entity in a context, extracting a speaker history and a visual context from the training data, selecting attributes of the target entity based on at least one of the speaker history, the visual context, and speaker preferences, generating a text expression referring to the target entity based on at least one of the selected attributes, the speaker history, and the context, and outputting the generated text expression. The weighted finite-state automaton can represent partial orderings of word pairs in the domain-specific training data. The weighted finite-state automaton can be speaker specific or speaker independent. The weighted finite-state automaton can include a set of weighted partial orderings of the training data for each possible realization.
摘要翻译： 本文公开了用于引用实体的系统，计算机实现的方法和有形的计算机可读介质。 该方法包括接收在上下文中描述目标实体的句子的特定领域的训练数据，从训练数据中提取讲者历史和视觉上下文，基于说话者的历史，视觉上的至少一个来选择目标实体的属性 上下文和说话人首选项，基于所选择的属性，说话者历史和上下文中的至少一个生成参考目标实体的文本表达，并输出所生成的文本表达。 加权有限状态自动机可以表示域特定训练数据中单词对的部分排序。 加权有限状态自动机可以是扬声器专用或扬声器独立的。 加权有限状态自动机可以包括用于每个可能实现的训练数据的一组加权部分排序。
-
公开(公告)号：US08990126B1
公开(公告)日：2015-03-24
申请号：US11462068
申请日：2006-08-03
申请人： Srinivas Bangalore , Alistair Conkie , Giuseppe Di Fabbrizio , Narendra Kumar Gupta , Amanda Joy Stent
发明人： Srinivas Bangalore , Alistair Conkie , Giuseppe Di Fabbrizio , Narendra Kumar Gupta , Amanda Joy Stent
CPC分类号： G06F17/211 , G06F17/27 , G06F17/30654
摘要： A method, system and computer readable medium that generates a dialog model for use in automated dialog is disclosed. The method may include collecting a plurality of task-oriented dialog interactions between users and human agents for a given domain, identifying one or more task in each dialog interaction, identifying one or more subtasks in each identified task and associating relations between the subtasks, identifying a dialog act and a set of predicate-argument relations for each subtask, generating one or more clauses from the set of predicate-argument relations, storing the tasks, subtasks, dialog acts predicate-argument relations, and clauses from each dialog interaction as a dialog interaction set, generating a dialog management model using the stored dialog interaction sets.
摘要翻译： 公开了一种生成用于自动对话中的对话模型的方法，系统和计算机可读介质。 该方法可以包括针对给定域收集用户和人类代理之间的多个面向任务的对话交互，识别每个对话交互中的一个或多个任务，识别每个被识别的任务中的一个或多个子任务并且关联子任务之间的关系，识别 每个子任务的对话行为和一组谓词 - 参数关系，从一组谓词参数关系生成一个或多个子句，将任务，子任务，对话动作谓词参数关系和每个对话框交互的子句存储为 对话交互集，使用存储的对话交互集生成对话管理模型。
-
6.发明申请Automatic Learning For Mapping Spoken/Text Descriptions Of Products Onto Available Products 失效自动学习将产品的口语/文字描述映射到可用的产品上公开(公告)号：US20070250404A1
公开(公告)日：2007-10-25
申请号：US11279802
申请日：2006-04-14
IPC分类号： G06Q30/00
CPC分类号： G06Q30/0603 , G06Q10/063 , G06Q10/06393 , G06Q10/087 , G06Q30/02 , G06Q40/08 , G06Q40/12
摘要： A method, processing device, and machine-readable medium are provided. Costs of states of a state space are calculated. Each state represent one or more available product attributes having zero or more decided attribute values. The calculating is based, at least in part, on training data associated with previously requested and offered products, determining a next state such that one or more products are available and a sum of values, including a cost of a next state and a cost of a perturbation of one of the one or more requested product attribute values to reach the next state is a minimum value. A value for a product attribute is mapped according to the minimum sum of values and product attribute values of available products.
摘要翻译： 提供了一种方法，处理装置和机器可读介质。 计算状态空间的状态成本。 每个状态表示具有零个或多个决定的属性值的一个或多个可用的产品属性。 该计算至少部分地基于与先前请求和提供的产品相关联的训练数据，确定下一个状态，使得一个或多个产品可用，并且包括下一个状态的成本和成本的值 一个或多个请求的产品属性值之一的扰动达到下一个状态是最小值。 产品属性的值根据可用产品的值和产品属性值的最小总和进行映射。
-
7.发明授权Automatic learning for mapping spoken/text descriptions of products onto available products 有权自动学习将产品的口语/文字描述映射到可用产品上公开(公告)号：US07937301B2
公开(公告)日：2011-05-03
申请号：US12632347
申请日：2009-12-07
CPC分类号： G06Q30/0603 , G06Q10/063 , G06Q10/06393 , G06Q10/087 , G06Q30/02 , G06Q40/08 , G06Q40/12
摘要： A method, processing device, and machine-readable medium are provided. Costs of states of a state space are calculated. Each state represent one or more available product attributes having zero or more decided attribute values. The calculating is based, at least in part, on training data associated with previously requested and offered products. Determining a next state such that one or more products are available and a sum of values, including a cost of a next state and a cost of a perturbation of one of the one or more requested product attribute values to reach the next state is a minimum value. A value for a product attribute is mapped according to the minimum sum of values and product attribute values of available products.
摘要翻译： 提供了一种方法，处理装置和机器可读介质。 计算状态空间的状态成本。 每个状态表示具有零个或多个决定的属性值的一个或多个可用的产品属性。 该计算至少部分地基于与先前请求和提供的产品相关联的训练数据。 确定下一状态，使得一个或多个产品可用，并且包括下一状态的成本和一个或多个所请求的产品属性值中的一个的扰动成本达到下一个状态的值的总和是最小的 值。 产品属性的值根据可用产品的值和产品属性值的最小总和进行映射。
-
8.发明授权Automatic learning for mapping spoken/text descriptions of products onto available products 失效自动学习将产品的口语/文字描述映射到可用产品上公开(公告)号：US07630917B2
公开(公告)日：2009-12-08
申请号：US11279802
申请日：2006-04-14
CPC分类号： G06Q30/0603 , G06Q10/063 , G06Q10/06393 , G06Q10/087 , G06Q30/02 , G06Q40/08 , G06Q40/12
摘要： A method, processing device, and machine-readable medium are provided. Costs of states of a state space are calculated. Each state represent one or more available product attributes having zero or more decided attribute values. The calculating is based, at least in part, on training data associated with previously requested and offered products. determining a next state such that one or more products are available and a sum of values, including a cost of a next state and a cost of a perturbation of one of the one or more requested product attribute values to reach the next state is a minimum value. A value for a product attribute is mapped according to the minimum sum of values and product attribute values of available products.
摘要翻译： 提供了一种方法，处理装置和机器可读介质。 计算状态空间的状态成本。 每个状态表示具有零个或多个决定的属性值的一个或多个可用的产品属性。 该计算至少部分地基于与先前请求和提供的产品相关联的训练数据。 确定下一状态使得一个或多个产品可用，并且包括下一个状态的成本和一个或多个所请求的产品属性值中的一个的扰动成本达到下一个状态的值的总和是最小的 值。 产品属性的值根据可用产品的值和产品属性值的最小总和进行映射。
-
公开(公告)号：US20100131274A1
公开(公告)日：2010-05-27
申请号：US12324340
申请日：2008-11-26
申请人： Amanda Stent , Srinivas Bangalore
发明人： Amanda Stent , Srinivas Bangalore
IPC分类号： G10L15/18
CPC分类号： G10L15/063 , G06F17/22 , G06F17/2241 , G06F17/227 , G10L15/005 , G10L15/04 , G10L15/08 , G10L15/18 , G10L15/183 , G10L15/22 , G10L25/12 , G10L2015/0638
摘要： Disclosed herein are systems, computer-implemented methods, and computer-readable media for dialog modeling. The method includes receiving spoken dialogs annotated to indicate dialog acts and task/subtask information, parsing the spoken dialogs with a hierarchical, parse-based dialog model which operates incrementally from left to right and which only analyzes a preceding dialog context to generate parsed spoken dialogs, and constructing a functional task structure of the parsed spoken dialogs. The method can further either interpret user utterances with the functional task structure of the parsed spoken dialogs or plan system responses to user utterances with the functional task structure of the parsed spoken dialogs. The parse-based dialog model can be a shift-reduce model, a start-complete model, or a connection path model.
摘要翻译： 本文公开了用于对话建模的系统，计算机实现的方法和计算机可读介质。 该方法包括接收注释以指示对话行为和任务/子任务信息的口头对话，用从层级，基于解析的对话模型解析口头对话，该对话模型从左向右逐渐操作，并且仅分析前一对话上下文以产生解析的口语对话 ，并构建解析的语音对话的功能任务结构。 该方法还可以用解析的口头对话的功能任务结构或用解析的口语对话的功能性任务结构对用户话语的计划系统响应来解释用户话语。 基于分析的对话模型可以是移位减少模型，起始完成模型或连接路径模型。
-
公开(公告)号：US09129601B2
公开(公告)日：2015-09-08
申请号：US12324340
申请日：2008-11-26
申请人： Amanda Stent , Srinivas Bangalore
发明人： Amanda Stent , Srinivas Bangalore
IPC分类号： G06F17/22 , G10L15/08 , G10L15/22 , G10L15/183 , G10L15/18
CPC分类号： G10L15/063 , G06F17/22 , G06F17/2241 , G06F17/227 , G10L15/005 , G10L15/04 , G10L15/08 , G10L15/18 , G10L15/183 , G10L15/22 , G10L25/12 , G10L2015/0638
摘要： Disclosed herein are systems, computer-implemented methods, and computer-readable media for dialog modeling. The method includes receiving spoken dialogs annotated to indicate dialog acts and task/subtask information, parsing the spoken dialogs with a hierarchical, parse-based dialog model which operates incrementally from left to right and which only analyzes a preceding dialog context to generate parsed spoken dialogs, and constructing a functional task structure of the parsed spoken dialogs. The method can further either interpret user utterances with the functional task structure of the parsed spoken dialogs or plan system responses to user utterances with the functional task structure of the parsed spoken dialogs. The parse-based dialog model can be a shift-reduce model, a start-complete model, or a connection path model.
摘要翻译： 本文公开了用于对话建模的系统，计算机实现的方法和计算机可读介质。 该方法包括接收注释以指示对话行为和任务/子任务信息的口头对话，用从层级，基于解析的对话模型解析口头对话，该对话模型从左向右逐渐操作，并且仅分析前一对话上下文以产生解析的口语对话 ，并构建解析的语音对话的功能任务结构。 该方法还可以用解析的口头对话的功能任务结构或用解析的口语对话的功能性任务结构对用户话语的计划系统响应来解释用户话语。 基于分析的对话模型可以是移位减少模型，起始完成模型或连接路径模型。
-
-
-
-
-
-
-
-
-
搜索结果
132 条记录 (耗时 0.026 秒)
