公开(公告)号：US08060369B2

公开(公告)日：2011-11-15

申请号：US12533300

申请日：2009-07-31

申请人： Srinivas Bangalore ,  Junlan Feng ,  Mazin G. Rahim

发明人： Srinivas Bangalore ,  Junlan Feng ,  Mazin G. Rahim

IPC分类号： G10L15/18 ,  G10L15/22

CPC分类号： G10L13/08 ,  G06F17/27 ,  G10L15/063 ,  G10L15/183 ,  G10L15/22 ,  H04M3/4936

摘要： Disclosed is a system and method for training a spoken dialog service component from website data. Spoken dialog service components typically include an automatic speech recognition module, a language understanding module, a dialog management module, a language generation module and a text-to-speech module. The method includes converting data from a structured database associated with a website to a structured text data set and a structured task knowledge base, extracting linguistic items from the structured database, and training a spoken dialog service component using at least one of the structured text data, the structured task knowledge base, or the linguistic items. The system includes modules configured to implement the method.

摘要翻译： 公开了一种用于从网站数据训练口语对话服务组件的系统和方法。 口语对话服务组件通常包括自动语音识别模块，语言理解模块，对话管理模块，语言生成模块和文本到语音模块。 该方法包括将来自与网站相关联的结构化数据库的数据转换为结构化文本数据集和结构化任务知识库，从结构化数据库中提取语言项目，以及使用至少一个结构化文本数据来训练口语对话服务组件 ，结构化任务知识库或语言项目。 该系统包括配置为实现该方法的模块。

公开(公告)号：US20100281435A1

公开(公告)日：2010-11-04

申请号：US12433320

申请日：2009-04-30

申请人： Srinivas Bangalore ,  Michael Johnston

发明人： Srinivas Bangalore ,  Michael Johnston

IPC分类号： G06F3/033

CPC分类号： G06F3/038 ,  G06F3/04883

摘要： Disclosed herein are systems, computer-implemented methods, and tangible computer-readable media for multimodal interaction. The method includes receiving a plurality of multimodal inputs associated with a query, the plurality of multimodal inputs including at least one gesture input, editing the at least one gesture input with a gesture edit machine. The method further includes responding to the query based on the edited gesture input and remaining multimodal inputs. The gesture inputs can be from a stylus, finger, mouse, and other pointing/gesture device. The gesture input can be unexpected or errorful. The gesture edit machine can perform actions such as deletion, substitution, insertion, and aggregation. The gesture edit machine can be modeled as a finite-state transducer. In one aspect, the method further includes generating a lattice for each input, generating an integrated lattice of combined meaning of the generated lattices, and responding to the query further based on the integrated lattice.

摘要翻译： 本文公开的是用于多模式交互的系统，计算机实现的方法和有形的计算机可读介质。 所述方法包括接收与查询相关联的多个多模式输入，所述多个多模式输入包括至少一个手势输入，用手势编辑机编辑所述至少一个手势输入。 该方法还包括基于编辑的手势输入和剩余的多模式输入来响应查询。 手势输入可以来自手写笔，手指，鼠标和其他指向/手势设备。 手势输入可能是意外或错误的。 手势编辑机可以执行删除，替换，插入和聚合等动作。 手势编辑机可以被建模为有限状态换能器。 一方面，所述方法还包括为每个输入生成格子，产生所产生的格子的组合意义的积分格，以及进一步基于所述积分格来响应所述查询。

公开(公告)号：US07783492B2

公开(公告)日：2010-08-24

申请号：US12179223

申请日：2008-07-24

申请人： Srinivas Bangalore ,  Michael J. Johnston

发明人： Srinivas Bangalore ,  Michael J. Johnston

IPC分类号： G10L21/00

CPC分类号： G06F3/017 ,  G01C21/3664 ,  G06F3/038 ,  G06F3/0481 ,  G06F3/04883 ,  G06F8/35 ,  G06F8/37 ,  G06K9/6885 ,  G10L15/183 ,  G10L15/193 ,  G10L15/24 ,  G10L15/32

摘要： Gesture and handwriting recognition agents provide possible interpretations of electronic ink. Recognition is performed on both individual strokes and combinations of strokes in the input ink lattice. The interpretations of electronic ink are classified and encoded as symbol complexes where symbols convey specific attributes of the contents of the stroke. The use of symbol complexes to represent strokes in the input ink lattice facilitates reference to sets of entities of a specific type.

摘要翻译： 手势和手写识别代理提供电子墨水的可能解释。 在输入墨水格子中的单个笔画和笔画的组合上执行识别。 电子墨水的解释被分类并编码为符号复合体，其中符号传达笔画内容的特定属性。 使用符号复合体来表示输入墨水格子中的笔画有助于参考特定类型的实体集合。

公开(公告)号：US20100131260A1

公开(公告)日：2010-05-27

申请号：US12324388

申请日：2008-11-26

申请人： Srinivas Bangalore ,  Vivek Kumar Rangarajan Sridhar

发明人： Srinivas Bangalore ,  Vivek Kumar Rangarajan Sridhar

IPC分类号： G06F17/28

CPC分类号： G06F17/28 ,  G06F17/279 ,  G06F17/289

摘要： Disclosed herein are systems, computer-implemented methods, and tangible computer-readable media for enriching spoken language translation with dialog acts. The method includes receiving a source speech signal, tagging dialog acts associated with the received source speech signal using a classification model, dialog acts being domain independent descriptions of an intended action a speaker carries out by uttering the source speech signal, producing an enriched hypothesis of the source speech signal incorporating the dialog act tags, and outputting a natural language response of the enriched hypothesis in a target language. Tags can be grouped into sets such as statement, acknowledgement, abandoned, agreement, question, appreciation, and other. The step of producing an enriched translation of the source speech signal uses a dialog act specific translation model containing a phrase translation table.

摘要翻译： 本文公开了系统，计算机实现的方法和有形计算机可读介质，用于通过对话行为丰富口语翻译。 该方法包括使用分类模型来接收源语音信号，与接收到的源语音信号相关联的标签对话动作，对话体是说话者通过发出源语音信号来执行的预期动作的域独立描述，产生丰富的假设 包含对话行为标签的源语音信号，并以目标语言输出丰富假说的自然语言响应。 标签可以分组，如声明，确认，放弃，协议，问题，升值等。 产生源语音信号的丰富翻译的步骤使用包含短语翻译表的对话行为特定翻译模型。

公开(公告)号：US20100098224A1

公开(公告)日：2010-04-22

申请号：US12644393

申请日：2009-12-22

申请人： Srinivas Bangalore ,  Mazin G. Rahim ,  Junlan Feng

发明人： Srinivas Bangalore ,  Mazin G. Rahim ,  Junlan Feng

IPC分类号： H04M1/64

CPC分类号： G06F17/243 ,  G06Q10/06311 ,  G06Q30/0201 ,  G06Q30/0635 ,  H04L12/6418 ,  H04M3/4938 ,  H04M7/0009 ,  H04M7/0036 ,  H04M2201/40 ,  H04M2201/60 ,  H04M2203/355

摘要： A system and method provides a natural language interface to world-wide web content. Either in advance or dynamically, webpage content is parsed using a parsing algorithm. A person using a telephone interface can provide speech information, which is converted to text and used to automatically fill in input fields on a webpage form. The form is then submitted to a database search and a response is generated. Information contained on the responsive webpage is extracted and converted to speech via a text-to-speech engine and communicated to the person.

摘要翻译： 系统和方法为世界各地的Web内容提供了一种自然语言界面。 提前或动态地，使用解析算法解析网页内容。 使用电话接口的人可以提供语音信息，其被转换成文本并用于自动填写网页表单上的输入字段。 然后将表单提交到数据库搜索，并生成响应。 包含在响应网页上的信息被提取并经由文本到语音引擎转换成语音，并传达给该人。

公开(公告)号：US07660828B2

公开(公告)日：2010-02-09

申请号：US11773659

申请日：2007-07-05

申请人： Michael J. Johnston ,  Srinivas Bangalore

发明人： Michael J. Johnston ,  Srinivas Bangalore

IPC分类号： G06F17/30

CPC分类号： G06F3/167 ,  G06F3/017 ,  G06F17/2247 ,  G06F17/2264 ,  G06F17/2775 ,  G06F17/28 ,  Y10S707/99934 ,  Y10S707/99943 ,  Y10S707/99948

摘要： When using finite-state devices to perform various functions, it is beneficial to use finite state devices representing regular grammars with terminals having markup-language-based semantics. By using markup-language-based symbols in the finite state devices, it is possible to generate valid markup-language expressions by concatenating the symbols representing the result of the performed function. The markup-language expression can be used by other applications and/or devices. Finite-state devices are used to convert strings of words and gestures into valid markup-language, for example, XML, expressions that can be used, for example, to provide an application program interface to underlying system applications.

摘要翻译： 当使用有限状态设备来执行各种功能时，使用表示具有基于标记语言的语义的终端的常规语法的有限状态设备是有益的。 通过在有限状态设备中使用基于标记语言的符号，可以通过连接表示所执行功能的结果的符号来生成有效的标记语言表达。 标记语言表达式可以由其他应用程序和/或设备使用。 有限状态设备用于将字串和手势转换为有效的标记语言，例如XML，可以使用的表达式，例如为基础系统应用程序提供应用程序接口。

公开(公告)号：US20090292529A1

公开(公告)日：2009-11-26

申请号：US12533300

申请日：2009-07-31

申请人： Srinivas Bangalore ,  Junlan Feng ,  Mazin G. Rahim

发明人： Srinivas Bangalore ,  Junlan Feng ,  Mazin G. Rahim

IPC分类号： G06F17/27 ,  G10L13/08

CPC分类号： G10L13/08 ,  G06F17/27 ,  G10L15/063 ,  G10L15/183 ,  G10L15/22 ,  H04M3/4936

摘要： Disclosed is a system and method for training a spoken dialog service component from website data. Spoken dialog service components typically include an automatic speech recognition module, a language understanding module, a dialog management module, a language generation module and a text-to-speech module. The method includes converting data from a structured database associated with a website to a structured text data set and a structured task knowledge base, extracting linguistic items from the structured database, and training a spoken dialog service component using at least one of the structured text data, the structured task knowledge base, or the linguistic items. The system includes modules configured to implement the method.

摘要翻译： 公开了一种用于从网站数据训练口语对话服务组件的系统和方法。 口语对话服务组件通常包括自动语音识别模块，语言理解模块，对话管理模块，语言生成模块和文本到语音模块。 该方法包括将来自与网站相关联的结构化数据库的数据转换为结构化文本数据集和结构化任务知识库，从结构化数据库中提取语言项目，以及使用至少一个结构化文本数据来训练口语对话服务组件 ，结构化任务知识库或语言项目。 该系统包括配置为实现该方法的模块。

公开(公告)号：US07451125B2

公开(公告)日：2008-11-11

申请号：US11268203

申请日：2005-11-07

申请人： Srinivas Bangalore

发明人： Srinivas Bangalore

IPC分类号： G06F17/00 ,  G06N5/02

CPC分类号： G06N5/025 ,  G06F17/27 ,  G06N5/02 ,  G06N99/005

摘要： A system, a method, and a machine-readable medium are provided. A group of linear rules and associated weights are provided as a result of machine learning. Each one of the group of linear rules is partitioned into a respective one of a group of types of rules. A respective transducer for each of the linear rules is compiled. A combined finite state transducer is created from a union of the respective transducers compiled from the linear rules.

摘要翻译： 提供了一种系统，方法和机器可读介质。 作为机器学习的结果，提供了一组线性规则和关联权重。 一组线性规则中的每一个被划分成一组类型的规则中的相应的一个。 编译每个线性规则的相应换能器。 从由线性规则编译的各个换能器的联合产生组合的有限状态换能器。

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。

公开(公告)号：US07373300B1

公开(公告)日：2008-05-13

申请号：US10740186

申请日：2003-12-18

申请人： Srinivas Bangalore ,  Junlan Feng ,  Mazin G. Rahim

发明人： Srinivas Bangalore ,  Junlan Feng ,  Mazin G. Rahim

IPC分类号： G10L15/22 ,  G10L13/08 ,  G10L15/18 ,  G06F17/27

CPC分类号： G10L13/08 ,  G06F17/27 ,  G10L15/063 ,  G10L15/183 ,  G10L15/22 ,  H04M3/4936

摘要： Disclosed is a system and method for generating a spoken dialog service from website data. Spoken dialog components typically include an automatic speech recognition module, a language understanding module, a dialog management module, a language generation module and a test-to-speech module. These components are capable of being automatically trained from processed website data. A website analyzer converts a website into structured text data set and a structured task knowledge base. The website analyzer further extracts linguistic items from the website data. The dialog components are automatically trained from the structured text data set, structured task knowledge base and linguistic items.

摘要翻译： 公开了一种用于从网站数据生成口语对话服务的系统和方法。 口语对话组件通常包括自动语音识别模块，语言理解模块，对话管理模块，语言生成模块和测试语音模块。 这些组件能够被处理的网站数据自动训练。 网站分析器将网站转换为结构化文本数据集和结构化任务知识库。 网站分析器进一步从网站数据中提取语言项目。 对话组件由结构化文本数据集，结构化任务知识库和语言项目自动进行训练。