利索-全球专利检索

  1. 登录
  2. 注册

搜索结果

444 条记录 (耗时 0.054 秒)

    Dialog Design Apparatus and Method
    42.
    发明申请
    Dialog Design Apparatus and Method 有权
    对话设计设备和方法

    公开(公告)号：US20100280819A1

    公开(公告)日：2010-11-04

    申请号：US12434394

    申请日：2009-05-01

    申请人： Inci Ozkaragoz ,  Yan Wang ,  Benjamin Ao

    发明人： Inci Ozkaragoz ,  Yan Wang ,  Benjamin Ao

    IPC分类号： G06F17/28

    CPC分类号： G06F17/289 ,  G06F8/38

    摘要： This invention relates to a dialog design apparatus and method. More specifically, this invention relates to a state oriented dialog design apparatus and method to facilitate the creation of natural language dialogs and creating data structures for voice user interfaces. The dialog design apparatus may include inputting means for receiving a user's prompt; response generating means for the user to generating at least one response; dialog structure generating means for structurally managing the user's input and response; and output means for outputting and displaying at least one dialog structure. A state in the present invention may include at least one system prompt and at least one response, and a linking unit may link a first state to a second state related to the first state, link the second state to a third state, and so on until certain system actions are achieved. A loop detecting unit in the present invention detects and identifies loops in the dialog structure.

    摘要翻译： 本发明涉及一种对话设计装置和方法。 更具体地，本发明涉及一种有助于自然语言对话的创建和为语音用户接口创建数据结构的状态对话设计装置和方法。 对话设计装置可以包括用于接收用户提示的输入装置; 响应产生装置，用于使用户产生至少一个响应; 对话结构产生装置，用于结构地管理用户的输入和响应; 以及用于输出和显示至少一个对话结构的输出装置。 本发明中的状态可以包括至少一个系统提示和至少一个响应，并且链接单元可以将第一状态链接到与第一状态相关的第二状态，将第二状态链接到第三状态等 直到达到某些系统动作。 本发明的环路检测单元检测并识别对话结构中的环路。

    Pipeline controller for context-based operation reconfigurable instruction set processor
    49.
    发明授权
    Pipeline controller for context-based operation reconfigurable instruction set processor 失效
    基于上下文操作的管道控制器可重构指令集处理器

    公开(公告)号：US07669042B2

    公开(公告)日：2010-02-23

    申请号：US11150427

    申请日：2005-06-10

    申请人： Eran Pisek ,  Jasmin Oz ,  Yan Wang

    发明人： Eran Pisek ,  Jasmin Oz ,  Yan Wang

    IPC分类号： G06F9/30

    CPC分类号： G06F8/4452 ,  G06F8/4432 ,  G06F9/381 ,  G06F9/4403 ,  Y02D10/41

    摘要： An instruction execution pipeline for use in a data processor. The instruction execution pipeline comprises: 1) an instruction fetch stage; 2) a decode stage; 3) an execution stage; and 4) a write-back stage. The instruction pipeline repetitively executes a loop of instructions by fetching and decoding a first instruction associated with the loop during a first iteration of the loop, storing first decoded instruction information associated with the first instruction during the first iteration of the loop, and using the stored first decoded instruction information during at least a second iteration of the loop without further fetching and decoding of the first instruction during the at least a second iteration of the loop.

    摘要翻译： 用于数据处理器的指令执行流水线。 指令执行流程包括：1）指令提取阶段; 2）解码阶段; 3）执行阶段; 和4）回写阶段。 所述指令流水线在所述循环的第一次迭代期间重复地执行指令循环，所述指令循环通过取出和解码与所述循环相关联的第一指令，在循环的第一次迭代期间存储与所述第一指令相关联的第一解码指令信息，并且使用所存储的 在循环的至少第二次迭代期间，在循环的至少第二次迭代期间，无需进一步获取和解码第一指令的第一解码指令信息。

    METHOD AND SYSTEM FOR OBTAINING PATH INFORMATION, PATH COMPUTATION ELEMENT
    50.
    发明申请
    METHOD AND SYSTEM FOR OBTAINING PATH INFORMATION, PATH COMPUTATION ELEMENT 有权
    获取路径信息的方法和系统，路径计算单元

    公开(公告)号：US20100039939A1

    公开(公告)日：2010-02-18

    申请号：US12603347

    申请日：2009-10-21

    申请人： Yan Wang

    发明人： Yan Wang

    IPC分类号： H04L12/56

    CPC分类号： H04L45/04 ,  H04L45/42

    摘要： A method and a system for obtaining path information as well as a PCE are disclosed herein. All the PCEs from the source PCE to the confluent PCE compute the shortest path from the source node to the egress border node of the domain covered by the respective PCE one by one, and send the computation result to the next-hop PCE until the confluent PCE. Meanwhile, all the PCEs from the destination PCE to the confluent PCE compute the shortest path from the destination node to the ingress border node of the domain covered by the respective PCE one by one, and send the computation result to the previous-hop PCE until the confluent PCE. After receiving computation results from the previous-hop PCE and the next-hop PCE, the confluent PCE computes the shortest path between the source node and the destination node, and sends the shortest path to the source PCE.

    摘要翻译： 本文公开了一种用于获得路径信息以及PCE的方法和系统。 从源PCE到汇合PCE的所有PCE计算从各个PCE逐个覆盖的域的源节点到出口边界节点的最短路径，并将计算结果发送到下一跳PCE直到汇合 PCE。 同时，从目的地PCE到汇合PCE的所有PCE逐个计算从目的地节点到相应PCE覆盖的域的入口边界节点的最短路径，并将计算结果发送到上一跳PCE直到 汇合的PCE。 汇聚PCE从前一跳PCE和下一跳PCE接收到计算结果后，计算源节点与目的节点之间的最短路径，并将最短路径发送到源PCE。