利索-全球专利检索

  1. 登录
  2. 注册

搜索结果

63 条记录 (耗时 0.025 秒)

    Grammar fragment acquisition using syntactic and semantic clustering
    1.
    发明授权
    Grammar fragment acquisition using syntactic and semantic clustering 有权
    使用语法和语义聚类的语法片段获取

    公开(公告)号：US06173261B2

    公开(公告)日：2001-01-09

    申请号：US09217635

    申请日：1998-12-21

    申请人： Kazuhiro Arai ,  Allen Louis Gorin ,  Giuseppe Riccardi ,  Jeremy Huntley Wright

    发明人： Kazuhiro Arai ,  Allen Louis Gorin ,  Giuseppe Riccardi ,  Jeremy Huntley Wright

    IPC分类号： G10L1500

    CPC分类号： G10L15/063 ,  G06F17/274 ,  G10L15/193 ,  G10L2015/0631

    摘要： A method and apparatus are provided for automatically acquiring grammar fragments for recognizing and understanding fluently spoken language. Grammar fragments representing a set of syntactically and semantically similar phrases may be generated using three probability distributions: of succeeding words, of preceding words, and of associated call-types. The similarity between phrases may be measured by applying Kullback-Leibler distance to these three probability distributions. Phrases being close in all three distances may be clustered into a grammar fragment.

    摘要翻译： 提供了一种方法和装置，用于自动获取用于识别和理解流利的口语的语法片段。 可以使用三个概率分布来生成代表一组语法和语义上类似的短语的语法片段：前一个单词的后续单词和相关联的呼叫类型。 可以通过将Kullback-Leibler距离应用于这三个概率分布来测量短语之间的相似性。 所有三个距离中的短语可能被聚集成语法片段。

    RECOGNIZING THE NUMERIC LANGUAGE IN NATURAL SPOKEN DIALOGUE
    2.
    发明申请
    RECOGNIZING THE NUMERIC LANGUAGE IN NATURAL SPOKEN DIALOGUE 有权
    识别自然语言对话中的数字语言

    公开(公告)号：US20120041763A1

    公开(公告)日：2012-02-16

    申请号：US13280884

    申请日：2011-10-25

    申请人： Mazin G. Rahim ,  Giuseppe Riccardi ,  Jeremy Huntley Wright ,  Bruce Melvin Buntschuh ,  Allen Louis Gorin

    发明人： Mazin G. Rahim ,  Giuseppe Riccardi ,  Jeremy Huntley Wright ,  Bruce Melvin Buntschuh ,  Allen Louis Gorin

    IPC分类号： G10L15/14

    CPC分类号： G10L15/142

    摘要： A system and a method are provided. A speech recognition processor receives unconstrained input speech and outputs a string of words. The speech recognition processor is based on a numeric language that represents a subset of a vocabulary. The subset includes a set of words identified as being for interpreting and understanding number strings. A numeric understanding processor contains classes of rules for converting the string of words into a sequence of digits. The speech recognition processor utilizes an acoustic model database. A validation database stores a set of valid sequences of digits. A string validation processor outputs validity information based on a comparison of a sequence of digits output by the numeric understanding processor with valid sequences of digits in the validation database.

    摘要翻译： 提供了一种系统和方法。 语音识别处理器接收无约束输入语音并输出一串字。 语音识别处理器基于代表词汇子集的数字语言。 该子集包括被识别为用于解释和理解数字串的一组单词。 数字理解处理器包含用于将字符串转换为数字序列的规则类型。 语音识别处理器利用声学模型数据库。 验证数据库存储一组有效的数字序列。 字符串验证处理器基于数字理解处理器输出的数字序列与验证数据库中的有效数字序列的比较来输出有效性信息。

    Recognizing the numeric language in natural spoken dialogue
    3.
    发明授权
    Recognizing the numeric language in natural spoken dialogue 有权
    认识到自然语言对话中的数字语言

    公开(公告)号：US08655658B2

    公开(公告)日：2014-02-18

    申请号：US13280884

    申请日：2011-10-25

    申请人： Mazin G. Rahim ,  Giuseppe Riccardi ,  Jeremy Huntley Wright ,  Bruce Melvin Buntschuh ,  Allen Louis Gorin

    发明人： Mazin G. Rahim ,  Giuseppe Riccardi ,  Jeremy Huntley Wright ,  Bruce Melvin Buntschuh ,  Allen Louis Gorin

    IPC分类号： G10L15/14 ,  G10L15/18

    CPC分类号： G10L15/142

    摘要： A system and a method are provided. A speech recognition processor receives unconstrained input speech and outputs a string of words. The speech recognition processor is based on a numeric language that represents a subset of a vocabulary. The subset includes a set of words identified as being for interpreting and understanding number strings. A numeric understanding processor contains classes of rules for converting the string of words into a sequence of digits. The speech recognition processor utilizes an acoustic model database. A validation database stores a set of valid sequences of digits. A string validation processor outputs validity information based on a comparison of a sequence of digits output by the numeric understanding processor with valid sequences of digits in the validation database.

    摘要翻译： 提供了一种系统和方法。 语音识别处理器接收无约束输入语音并输出一串字。 语音识别处理器基于代表词汇子集的数字语言。 该子集包括被识别为用于解释和理解数字串的一组单词。 数字理解处理器包含用于将字符串转换为数字序列的规则类型。 语音识别处理器利用声学模型数据库。 验证数据库存储一组有效的数字序列。 字符串验证处理器基于数字理解处理器输出的数字序列与验证数据库中的有效数字序列的比较来输出有效性信息。

    Method and system for automatic detecting morphemes in a task classification system using lattices
    4.
    发明授权
    Method and system for automatic detecting morphemes in a task classification system using lattices 有权
    在使用格子的任务分类系统中自动检测语素的方法和系统

    公开(公告)号：US07620548B2

    公开(公告)日：2009-11-17

    申请号：US11854720

    申请日：2007-09-13

    申请人： Allen Louis Gorin ,  Dijana Petrovska-Delacretaz ,  Giuseppe Riccardi ,  Jeremy Huntley Wright

    发明人： Allen Louis Gorin ,  Dijana Petrovska-Delacretaz ,  Giuseppe Riccardi ,  Jeremy Huntley Wright

    IPC分类号： G10L15/06

    CPC分类号： G10L15/08

    摘要： The invention concerns a method and system for detecting morphemes in a user's communication. The method may include recognizing a lattice of phone strings from the user's input communication, the lattice representing a distribution over the phone strings, and detecting morphemes in the user's input communication using the lattice. The morphemes may be acoustic and/or non-acoustic. The morphemes may represent any unit or sub-unit of communication including phones, diphones, phone-phrases, syllables, grammars, words, gestures, tablet strokes, body movements, mouse clicks, etc. The training speech may be verbal, non-verbal, a combination of verbal and non-verbal, or multimodal.

    摘要翻译： 本发明涉及用于检测用户通信中的语素的方法和系统。 该方法可以包括从用户的输入通信识别电话串的格子，格子表示电话串上的分布，以及使用网格检测用户的输入通信中的语素。 语素可以是声学和/或非声学的。 语素可以代表通信的任何单位或子单位，包括手机，双耳，电话短语，音节，语法，单词，手势，平板笔画，身体动作，鼠标点击等。训练语言可以是口头上，非言语的 ，口头和非言语或多式联运。

    Method for task classification using morphemes
    9.
    发明授权
    Method for task classification using morphemes 有权
    使用语素进行任务分类的方法

    公开(公告)号：US07085720B1

    公开(公告)日：2006-08-01

    申请号：US09690721

    申请日：2000-10-18

    申请人： Allen Louis Gorin ,  Dijana Petrovska-Delacretaz ,  Giuseppe Riccardi ,  Jeremy Huntley Wright

    发明人： Allen Louis Gorin ,  Dijana Petrovska-Delacretaz ,  Giuseppe Riccardi ,  Jeremy Huntley Wright

    IPC分类号： G10L15/18

    CPC分类号： G10L15/1822 ,  G10L15/02 ,  G10L15/1815 ,  G10L15/24

    摘要： The invention concerns a method of task classification using morphemes which operates on the task objective of a user. The morphemes may be generated by clustering selected ones of the salient sub-morphemes selected from training speech which are semantically and syntactically similar. The method may include detecting morphemes present in the user's input communication, and making task-type classification decisions based on the detected morphemes in the user's input communication. The morphemes may be verbal and/or non-verbal.

    摘要翻译： 本发明涉及使用对用户的任务目标进行操作的语素的任务分类方法。 语素可以通过聚类从语义和语法上相似的训练语音中选出的突出的子语素中产生。 该方法可以包括检测用户输入通信中存在的语素，并且基于用户输入通信中检测到的语素来进行任务类型分类决定。 语素可能是口头和/或非言语。

    Method for generating morphemes
    10.
    发明授权
    Method for generating morphemes 有权
    生成语素的方法

    公开(公告)号：US06681206B1

    公开(公告)日：2004-01-20

    申请号：US09690903

    申请日：2000-10-18

    申请人： Allen Louis Gorin ,  Dijana Petrovska-Delacretaz ,  Giuseppe Riccardi ,  Jeremy Huntley Wright

    发明人： Allen Louis Gorin ,  Dijana Petrovska-Delacretaz ,  Giuseppe Riccardi ,  Jeremy Huntley Wright

    IPC分类号： G10L1506

    CPC分类号： G06F17/2755

    摘要： The invention concerns a method of generating morphemes for speech recognition and understanding. The method may include receiving training speech, selecting candidate sub-morphemes from the training speech, selecting salient sub-morphemes from the candidate sub-morphemes based on salience measurements, and clustering the salient sub-morphemes based on semantic and syntactic similarities into morphemes. The morphemes may be acoustic and/or non-acoustic. The sub-morphemes may represent any sub-unit of communication including phones, phone-phrases, grammars, diphones, words, gestures, tablet strokes, body movements, mouse clicks, etc. The training speech may be verbal, non-verbal, a combination of verbal and non-verbal, or multimodal.

    摘要翻译： 本发明涉及一种产生用于语音识别和理解的语素的方法。 该方法可以包括接收训练语音，从训练语音中选择候选子语素，基于显着性测量从候选子语素中选择突出的子语素，并将基于语义和句法相似性的突出子语素聚类成语素。 语素可以是声学和/或非声学的。 子语素可以代表任何通信的子单元，包括手机，电话短语，语法，双音，单词，手势，平板笔画，身体动作，鼠标点击等。训练语音可以是口头上，非口头的，一个 口头和非言语或多式联运。