中科微点-全球专利检索

  1. Login
  2. Sign Up

Search Results

27 records (took 0.016 seconds)

    PREDICTIVE 411
    1.
    发明申请
    PREDICTIVE 411 审中-公开
    预测411

    公开(公告)号：WO2013169912A3

    公开(公告)日：2014-01-23

    申请号：PCT/US2013040160

    申请日：2013-05-08

    Applicant: 24 7 CUSTOMER INC ,  NGUYEN PATRICK ,  LINDSETH TREY ,  NGUYEN QUANG

    Inventor： NGUYEN PATRICK ,  LINDSETH TREY ,  NGUYEN QUANG

    IPC: G06F17/30

    CPC classification number: G06F17/30389 ,  G06F17/30477 ,  G06F17/30528 ,  G06F17/3053 ,  G06F17/30554 ,  G06F17/3056 ,  G06F17/3097 ,  G06Q10/02 ,  G06Q10/10 ,  G06Q10/109 ,  G06Q10/1095 ,  G06Q30/02

    Abstract: A system predicts the intent of a user and proactively offers to perform a query that satisfies that intent. Upon the user's acceptance of the offer, the system begins a search for related information. The system examines such factors as search terms typed or spoken by said user, historical attributes of said user, historical journey attributes of said user, current journey attributes of said user, user location, user movement, current time, user profile, user calendar, user information stored on, or associated with, a device within the user's possession. The system then makes a prediction of any of the user's intent, query category, and issue category. Based upon the results of the system's prediction, a query that is relevant to the user's intent and/or issue categories is presented and, upon the user's command, the results of the search are returned to the user.

    Abstract translation: 系统预测用户的意图并主动提供执行满足该意图的查询。 在用户接受报价后，系统开始搜索相关信息。 该系统检查所述用户输入或说出的搜索词，所述用户的历史属性，所述用户的历史旅程属性，所述用户的当前旅程属性，用户位置，用户移动，当前时间，用户简档，用户日历， 存储在用户拥有的设备上或与之相关联的用户信息。 系统然后对用户的意图，查询类别和问题类别进行预测。 基于系统预测的结果，呈现与用户意图和/或问题类别相关的查询，并且根据用户的命令将搜索结果返回给用户。

    PATTERN MATCHING FOR LARGE VOCABULARY SPEECH RECOGNITION WITH PACKED DISTRIBUTION AND LOCALIZED TRELLIS ACCESS
    4.
    发明申请
    PATTERN MATCHING FOR LARGE VOCABULARY SPEECH RECOGNITION WITH PACKED DISTRIBUTION AND LOCALIZED TRELLIS ACCESS 审中-公开

    公开(公告)号：WO2003090203A3

    公开(公告)日：2003-10-30

    申请号：PCT/US2003/008063

    申请日：2003-03-19

    Applicant: MATSUSHITA ELECTRIC INDUSTRIAL CO., LTD. ,  RIGAZIO, Luca ,  NGUYEN, Patrick

    Inventor： RIGAZIO, Luca ,  NGUYEN, Patrick

    IPC: G10L15/00

    Abstract: A method is provided for improving pattern matching in a speech recognition system having a plurality of acoustic models (20). Similarity measures for acoustic feature vectors (54) are determined in groups that are then buffered into cache memory (59). To further reduce computational processing, the acoustic data may be partitioned amongst a plurality of processing nodes (66, 67, 68). In addition, a priori knowledge of the spoken order may be used to establish the access order (124) used to copy records from the main speech parameter table (120, 200) into a sub-table (130, 204). The sub-table is processed such that the entries are in contiguous memory locations (206) and sorted according to the processing order (208). The speech processing algorithm is then directed to operate upon the sub-table (210) which causes the processor to load the sub-table into high speed cache memory (104, 212).

    PATTERN MATCHING FOR LARGE VOCABULARY SPEECH RECOGNITION WITH PACKED DISTRIBUTION AND LOCALIZED TRELLIS ACCESS
    5.
    发明申请
    PATTERN MATCHING FOR LARGE VOCABULARY SPEECH RECOGNITION WITH PACKED DISTRIBUTION AND LOCALIZED TRELLIS ACCESS 审中-公开
    大规模语音识别与包装分发和本地化TRELLIS访问的模式匹配

    公开(公告)号：WO2003090203A2

    公开(公告)日：2003-10-30

    申请号：PCT/US2003/008063

    申请日：2003-03-19

    Applicant: MATSUSHITA ELECTRIC INDUSTRIAL CO., LTD. ,  RIGAZIO, Luca ,  NGUYEN, Patrick

    Inventor： RIGAZIO, Luca ,  NGUYEN, Patrick

    IPC: G10L

    CPC classification number: G10L15/08 ,  G10L15/10 ,  G10L15/285 ,  G10L15/30 ,  G10L15/34

    Abstract: A method is provided for improving pattern matching in a speech recognition system having a plurality of acoustic models (20). Similarity measures for acoustic feature vectors (54) are determined in groups that are then buffered into cache memory (59). To further reduce computational processing, the acoustic data may be partitioned amongst a plurality of processing nodes (66, 67, 68). In addition, a priori knowledge of the spoken order may be used to establish the access order (124) used to copy records from the main speech parameter table (120, 200) into a sub-table (130, 204). The sub-table is processed such that the entries are in contiguous memory locations (206) and sorted according to the processing order (208). The speech processing algorithm is then directed to operate upon the sub-table (210) which causes the processor to load the sub-table into high speed cache memory (104, 212).

    Abstract translation: 提供了一种用于改进具有多个声学模型（20）的语音识别系统中的模式匹配的方法。 以随后缓冲到高速缓存存储器（59）中的组确定声学特征向量（54）的相似性度量。 为了进一步减少计算处理，可以在多个处理节点（66,67,68）之间划分声学数据。 此外，可以使用口语顺序的先验知识来建立用于将记录从主语音参数表（120,200）复制到子表（130,204）中的访问顺序（124）。 处理子表使得条目在连续存储器位置（206）中并根据处理顺序（208）进行排序。 语音处理算法随后被引导以对子表（210）进行操作，这使得处理器将子表加载到高速缓存存储器（104,212）中。

    VOICE PERSONALIZATION OF SPEECH SYNTHESIZER
    6.
    发明申请
    VOICE PERSONALIZATION OF SPEECH SYNTHESIZER 审中-公开
    语音合成器的语音个性化

    公开(公告)号：WO2002069323A1

    公开(公告)日：2002-09-06

    申请号：PCT/US2002/005631

    申请日：2002-02-25

    Applicant: MATSUSHITA ELECTRIC INDUSTRIAL CO., LTD. ,  JUNQUA, Jean-Claude ,  PERRONNIN, Florent ,  KUHN, Roland ,  NGUYEN, Patrick

    Inventor： JUNQUA, Jean-Claude ,  PERRONNIN, Florent ,  KUHN, Roland ,  NGUYEN, Patrick

    IPC: G10L13/02

    CPC classification number: G10L13/04 ,  G10L2021/0135

    Abstract: The speech synthesizer is personalized to sound like or mimic the speech characteristics of an individual speaker. The individual speaker provides a quantity of enrollment data (18), which can be extracted from a short quantity of speech, and the system modifies the base synthesis parameters (12) to more closely resemble those of the new speaker (36). More specifically, the synthesis parameters (12) may be decomposed into speaker dependent parameters (30), such as context-independent parameters, and speaker independent parameters (32), such as contextindependent parameters, and speaker independent parameters (32), such as context dependent parameters. The speaker dependent parameters (30) are adapted using enrollment data (18) from the new speaker. After adaptation, the speaker dependent parameters (30) are combined with the speaker independent parameters (32) to provide a set of personalized synthesis parameters (42).

    Abstract translation: 语音合成器被个性化以发音或模仿单个扬声器的语音特征。 单个扬声器提供一定量的登记数据（18），其可以从短语言提取，并且系统将基本合成参数（12）修改为更接近于新的扬声器（36）的参考数据。 更具体地，合成参数（12）可以被分解为与扬声器相关的参数（30），诸如与上下文无关的参数，以及与扬声器无关的参数（32），诸如上下文相关参数和与扬声器无关的参数（32），诸如 上下文相关参数。 使用来自新说话者的登记数据（18）来调整说话者依赖参数（30）。 在适应之后，将扬声器相关参数（30）与扬声器独立参数（32）组合以提供一组个性化合成参数（42）。

    MEDIA PRODUCTION SYSTEM USING TIME ALIGNMENT TO SCRIPTS
    10.
    发明申请
    MEDIA PRODUCTION SYSTEM USING TIME ALIGNMENT TO SCRIPTS 审中-公开
    媒体生产系统使用时间对齐到脚本

    公开(公告)号：WO2005094336A2

    公开(公告)日：2005-10-13

    申请号：PCT/US2005010477

    申请日：2005-03-29

    Applicant: MATSUSHITA ELECTRIC IND CO LTD ,  BOMAN ROBERT C ,  NGUYEN PATRICK ,  JUNQUA JEAN-CLAUDE

    Inventor： BOMAN ROBERT C ,  NGUYEN PATRICK ,  JUNQUA JEAN-CLAUDE

    IPC: G10L15/26

    CPC classification number: G10L15/26

    Abstract: A media production system includes a textual alignment module aligning multiple speech recordings to textual lines of a script based on speech recognition results. A navigation module responds to user navigation selections respective of the textual lines of the script by communicating to the user corresponding, line-specific portions of the multiple speech recordings. An editing module responds to user associations of multiple speech recordings with textual lines by accumulating line-specific portions of the multiple speech recordings in a combination recording based on at least one of relationships of textual lines in the script to the combination recording, and temporal alignments between the multiple speech recordings and the combination recording.

    Abstract translation: 媒体制作系统包括文本对准模块，其基于语音识别结果将多个语音记录与脚本的文本行对齐。 导航模块通过与用户对应的多个语音记录的线特定部分通信来响应相应于脚本的文本行的用户导航选择。 编辑模块通过基于脚本中的文本行的关系与组合记录中的至少一种相结合记录来组合记录中的多个语音记录的行特定部分来累积多个语音记录与文本行的响应，以及时间对齐 在多个语音记录和组合记录之间。

Patent Agency Ranking