Biasing a speech recognizer based on prompt context
    1.
    发明授权
    Biasing a speech recognizer based on prompt context 有权
    基于提示语言偏好语音识别器

    公开(公告)号:US07542907B2

    公开(公告)日:2009-06-02

    申请号:US10741465

    申请日:2003-12-19

    IPC分类号: G10L15/22 G10L15/14

    CPC分类号: G10L15/183 G10L15/1815

    摘要: A method, a system, and an apparatus biasing a speech recognizer based on prompt context. The present invention is capable of analyzing the words used in the prompt given to the user. Then, a set of words the user is likely to say in response to the prompt is determined. The word set may be determined using a technology used by the speech recognition system, such as n-grams, grammars, or both. The speech recognition system boosts the probabilities of the analyzed words in the word set by a preconfigured amount. The preconfigured amount is selected based on collected data.

    摘要翻译: 一种基于快速上下文偏好语音识别器的方法,系统和装置。 本发明能够分析给予用户的提示中使用的单词。 然后,确定用户响应于提示可能说的一组单词。 可以使用语音识别系统所使用的技术,例如n-gram,语法,或两者​​来确定该单词集合。 语音识别系统通过预配置的量来提高词组中分析词的概率。 基于收集的数据选择预配置的数量。

    Semantic language modeling and confidence measurement
    2.
    发明授权
    Semantic language modeling and confidence measurement 有权
    语义语言建模和置信度测量

    公开(公告)号:US07475015B2

    公开(公告)日:2009-01-06

    申请号:US10655838

    申请日:2003-09-05

    IPC分类号: G10L15/00

    CPC分类号: G10L15/1815

    摘要: A system and method for speech recognition includes generating a set of likely hypotheses in recognizing speech, rescoring the likely hypotheses by using semantic content by employing semantic structured language models, and scoring parse trees to identify a best sentence according to the sentence's parse tree by employing the semantic structured language models to clarify the recognized speech.

    摘要翻译: 一种用于语音识别的系统和方法包括在识别语音中产生一组可能的假设,通过使用语义结构化语言模型通过使用语义内容来重新计算可能的假设,并且通过采用语法结构语言模型对解析树进行评分以识别根据句子的解析树的最佳句子 语义结构语言模型来澄清公认的言语。

    Including statistical NLU models within a statistical parser
    3.
    发明授权
    Including statistical NLU models within a statistical parser 有权
    在统计解析器中包括统计NLU模型

    公开(公告)号:US06963831B1

    公开(公告)日:2005-11-08

    申请号:US09696803

    申请日:2000-10-25

    申请人: Mark E. Epstein

    发明人: Mark E. Epstein

    IPC分类号: G06F17/20 G06F17/27

    CPC分类号: G06F17/2715

    摘要: A method and system for statistical parsing. The method involves a series of steps. The system can apply a statistical natural language understanding (NLU) model to test input for identifying substrings within the text input. The statistical NLU model can be selected for identifying a particular class of substring. The system can examine each identified substring using an inventory of queries corresponding to the reusable statistical NLU model.

    摘要翻译: 一种用于统计解析的方法和系统。 该方法涉及一系列步骤。 该系统可以应用统计自然语言理解(NLU)模型来测试输入以识别文本输入内的子串。 可以选择统计NLU模型来识别特定类别的子串。 系统可以使用与可重用的统计NLU模型相对应的查询清单来检查每个确定的子字符串。

    Method of tuning a decision network and a decision tree model
    4.
    发明授权
    Method of tuning a decision network and a decision tree model 失效
    调整决策网络和决策树模型的方法

    公开(公告)号:US06889219B2

    公开(公告)日:2005-05-03

    申请号:US10054051

    申请日:2002-01-22

    IPC分类号: G06N5/02 G06F17/00 G06N5/00

    CPC分类号: G06N5/025

    摘要: A method of tuning a decision network can include a series of steps. The steps can include identifying a deviation between a correct interpretation of a data item and an incorrect interpretation of the data item. In a decision network including a hierarchical set of nodes and leaves, a path of traversed nodes in the decision network resulting in the deviation can be determined. The nodes can correspond to queries. A measure of goodness at a node in the path can be calculated using at least one new query as a replacement for an existing query to determine whether the measure of goodness improves using the new query compared to the old query. If the measure of goodness improves, at least one new query can be selected and the decision network can be regrown from the node down through the leaves using the at least one new query at the node.

    摘要翻译: 调整决策网络的方法可以包括一系列步骤。 这些步骤可以包括识别数据项的正确解释与数据项的不正确解释之间的偏差。 在包括节点和树叶的分级集合的决策网络中,可以确定导致偏差的决策网络中遍历节点的路径。 节点可以对应于查询。 可以使用至少一个新查询作为现有查询的替代来计算路径中的节点的良好度量,以确定与旧查询相比,使用新查询是否改善了度量。 如果善意度提高,则可以选择至少一个新查询,并且可以使用该节点处的至少一个新查询从节点向下再循环决策网络。

    Method of identifying members of classes in a natural language understanding system
    5.
    发明授权
    Method of identifying members of classes in a natural language understanding system 失效
    在自然语言理解系统中识别课程成员的方法

    公开(公告)号:US06735560B1

    公开(公告)日:2004-05-11

    申请号:US09773137

    申请日:2001-01-31

    申请人: Mark E. Epstein

    发明人: Mark E. Epstein

    IPC分类号: G06F1727

    CPC分类号: G06F17/278 G06F17/2775

    摘要: A method of configuring classes in a natural language understanding (NLU) system. The method can include the steps of assigning a unique value to members of a class in the NLU system. The step of generating possible substrings from the members in the class also can be included. Additionally, for each generated substring having at least one term in common with one of the members in the class, the step of associating with the generated substring the unique value assigned to the member can be included.

    摘要翻译: 一种在自然语言理解(NLU)系统中配置课程的方法。 该方法可以包括为NLU系统中的类的成员分配唯一值的步骤。 也可以包括从类中的成员生成可能的子字符串的步骤。 此外,对于每个生成的子字符串,其具有至少一个与类中的成员之一共同的项,可以包括与生成的子串相关联的步骤,分配给该成员的唯一值。

    Method and apparatus for embedding grammars in a natural language understanding (NLU) statistical parser
    6.
    发明授权
    Method and apparatus for embedding grammars in a natural language understanding (NLU) statistical parser 失效
    在自然语言理解(NLU)统计解析器中嵌入语法的方法和装置

    公开(公告)号:US06983239B1

    公开(公告)日:2006-01-03

    申请号:US09696764

    申请日:2000-10-25

    申请人: Mark E. Epstein

    发明人: Mark E. Epstein

    IPC分类号: G06F17/27

    摘要: A method and system for use in a natural language understanding system for including grammars within a statistical parser. The method involves a series of steps. The invention receives a text input. The invention applies a first context free grammar to the text input to determine substrings and corresponding parse trees, wherein the substrings and corresponding parse trees further correspond to the first context free grammar. Additionally, the invention can examine each possible substring using an inventory of queries corresponding to the CFG.

    摘要翻译: 用于在统计解析器中包含语法的自然语言理解系统中使用的方法和系统。 该方法涉及一系列步骤。 本发明接收文本输入。 本发明将第一上下文无关语法应用于文本输入以确定子串和相应的解析树,其中所述子串和相应的解析树进一步对应于第一语境自由语法。 此外,本发明可以使用与CFG相对应的查询的清单来检查每个可能的子串。

    Method and apparatus for providing multiple output channels in a microphone
    7.
    发明授权
    Method and apparatus for providing multiple output channels in a microphone 失效
    用于在麦克风中提供多个输出通道的方法和装置

    公开(公告)号:US06959095B2

    公开(公告)日:2005-10-25

    申请号:US09927690

    申请日:2001-08-10

    IPC分类号: H04R3/00

    CPC分类号: H04R3/00

    摘要: Methods and apparatus for providing multiple output channels in a microphone. More particularly, provision is made for an arrangement wherein a single microphone is adapted to produce one or more different audio outputs depending upon characteristics of a speaker or user of the microphone while facilitating a high degree of accuracy in the recognition of the user or speaker by the arrangement. The microphone is adapted to produce one or more different audio streams or outputs depending upon the speaker presently using the microphone. In effect, this can be readily implemented by a main user or speaker, such as an interviewer on a radio or TV talk show, or any speaker in a conference room, intending to control the audio output streams by suitably activating a button or switch.

    摘要翻译: 用于在麦克风中提供多个输出通道的方法和装置。 更具体地,提供了一种布置,其中单个麦克风适于根据麦克风的扬声器或用户的特性产生一个或多个不同的音频输出,同时便于用户或扬声器的识别中的高度准确度 安排。 麦克风适于根据目前使用麦克风的扬声器产生一个或多个不同的音频流或输出。 实际上,这可以容易地由诸如无线电或电视谈话节目上的访问者或会议室中的任何扬声器的主要用户或扬声器实现,旨在通过适当地激活按钮或开关来控制音频输出流。

    Creating a hierarchical tree of language models for a dialog system based on prompt and dialog context
    8.
    发明授权
    Creating a hierarchical tree of language models for a dialog system based on prompt and dialog context 有权
    基于提示和对话框上下文创建对话系统的语言模型层次树

    公开(公告)号:US06754626B2

    公开(公告)日:2004-06-22

    申请号:US09798655

    申请日:2001-03-01

    申请人: Mark E. Epstein

    发明人: Mark E. Epstein

    IPC分类号: G10L1526

    CPC分类号: G10L15/197 G10L15/183

    摘要: The invention disclosed herein concerns a method of converting speech to text using a hierarchy of contextual models. The hierarchy of contextual models can be statistically smoothed into a language model. The method can include processing text with a plurality of contextual models. Each one of the plurality of contextual models can correspond to a node in a hierarchy of the plurality of contextual models. Also included can be identifying at least one of the contextual models relating to the text and processing subsequent user spoken utterances with the identified at least one contextual model.

    摘要翻译: 本文公开的发明涉及使用上下文模型的层次将语音转换为文本的方法。 语境模型的层次结构可以被统计学平滑化为语言模型。 该方法可以包括用多个上下文模型处理文本。 多个上下文模型中的每一个可以对应于多个上下文模型的层次结构中的节点。 还包括可以识别与文本相关的上下文模型中的至少一个以及使用所识别的至少一个上下文模型来处理随后的用户口语话语。

    Bi-directional natural language system for interfacing with multiple back-end applications
    9.
    发明授权
    Bi-directional natural language system for interfacing with multiple back-end applications 有权
    用于与多个后端应用程序进行连接的双向自然语言系统

    公开(公告)号:US06442522B1

    公开(公告)日:2002-08-27

    申请号:US09415660

    申请日:1999-10-12

    IPC分类号: G10L1518

    摘要: A system and method for servicing natural language requests with a plurality of remote host systems. The system utilizes a computer program that comprises: (1) an input system for inputting an NL command; (2) a translation system that extracts a request from the NL command and stores the request in a host-independent format; and (3) a routing system for servicing the request, wherein the routing system comprises a mechanism for selecting a host, for converting the request into a host dependent directive, and for forwarding the directive to the selected host. The system may further include a speech recognition system, a local data source for servicing the NL command, templates for converting the request into the host dependent directive, a heuristic for selecting the host, and an output system for obtaining and outputting the response.

    摘要翻译: 一种用于利用多个远程主机系统来服务自然语言请求的系统和方法。 该系统利用计算机程序,其包括:(1)用于输入NL命令的输入系统; (2)翻译系统,其从NL命令中提取请求并将请求存储在与主机无关的格式中; 以及(3)用于对所述请求进行服务的路由系统,其中所述路由系统包括用于选择主机的机制,用于将所述请求转换为主机相关指令,以及将所述指令转发到所选主机。 该系统还可以包括语音识别系统,用于服务NL命令的本地数据源,用于将请求转换为主机相关指令的模板,用于选择主机的启发式,以及用于获得和输出响应的输出系统。

    Cleaning composition and method for utilizing same
    10.
    发明授权
    Cleaning composition and method for utilizing same 失效
    清洗组合物及其使用方法

    公开(公告)号:US5415811A

    公开(公告)日:1995-05-16

    申请号:US193564

    申请日:1994-02-08

    摘要: An aqueous cleaning composition useful for cleaning hard surfaces such as glass, and a method for utilizing same, wherein relatively no residue is formed by the composition upon application to the surface. The composition comprises an aqueous mixture of an alcohol, a glycol ether, and a surfactant, wherein the surfactant is a fluorosurfactant present in an amount equal to or less than 40 parts per million, preferably 30 parts per million, based on the total composition.

    摘要翻译: 用于清洁硬表面如玻璃的水性清洁组合物及其使用方法,其中组合物在施用于表面时相对没有形成残留物。 该组合物包含醇,二醇醚和表面活性剂的水性混合物,其中表面活性剂是基于总组合物存在的含量等于或小于40ppm,优选30ppm的含氟表面活性剂。