公开(公告)号：US20050071674A1

公开(公告)日：2005-03-31

申请号：US10674285

申请日：2003-09-29

申请人： Wu Chou ,  Juan Li ,  Feng Liu ,  Xueshan Shan

发明人： Wu Chou ,  Juan Li ,  Feng Liu ,  Xueshan Shan

IPC分类号： H04L12/56 ,  H04L29/06 ,  H04L29/08 ,  G06F15/16

CPC分类号： H04L63/02 ,  H04L29/06027 ,  H04L63/08 ,  H04L65/1063 ,  H04L65/1073 ,  H04L67/2823 ,  H04L67/2857 ,  H04L69/329 ,  H04W12/06 ,  H04W12/08

摘要： A method and apparatus are provided for the secure delivery of converged services to users of wireless devices in a wireless environment. An application layer broker is position between the user devices and the application server to provide an indirect coupling between the enterprise application server and the wireless user devices. The application layer broker links the wireless environment to the enterprise application server through an event triggered content delivery mechanism without providing a direct link between the wireless user device and the enterprise application server. Delivered content can only be accessed by the intended user and the authorized device. The application layer broker provides adaptation to various user devices and the various capabilities associated with each user device. Wireless devices can be upgraded independently from upgrades to the application server. The decoupling performed by the application layer broker separates the service control, such as user registration from the service delivery.

摘要翻译： 提供了一种用于在无线环境中将融合服务安全传递给无线设备的用户的方法和装置。 应用层代理是用户设备和应用服务器之间的位置，以提供企业应用服务器与无线用户设备之间的间接耦合。 应用层代理通过事件触发的内容递送机制将无线环境链接到企业应用服务器，而不在无线用户设备和企业应用服务器之间提供直接链路。 交付的内容只能由目标用户和授权设备访问。 应用层代理提供对各种用户设备的适应以及与每个用户设备相关联的各种能力。 无线设备可以独立升级到应用服务器。 应用层代理执行的解耦将服务控制分离，例如用户注册与服务传递。

公开(公告)号：US6006181A

公开(公告)日：1999-12-21

申请号：US928372

申请日：1997-09-12

申请人： Eric Rolse Buhrke ,  Wu Chou

发明人： Eric Rolse Buhrke ,  Wu Chou

IPC分类号： G10L5/06 ,  G10L9/00

CPC分类号： G10L15/30

摘要： A continuous speech decoder that is built up of multiple layers. Each of the layers uses independent knowledge sources and rules, but all the layers cooperate to quickly decode the speech input into words. A first layer is concerned with acoustic data, a second layer with phone data of speech and a third layer concerns word data and word sequences. By separating these layers, the higher layers can be made time independent and asynchronous. Thus the asynchronous layers can process data quickly and give fast support to the first layer which keeps a dynamic record called a dynamic network of the most likely continuous speech results. The speed and separation of this decoder allows better memory efficiency and better decoder results compared to previously known continuous speech decoders.

摘要翻译： 连续语音解码器，由多层构成。 每个层使用独立的知识源和规则，但是所有的层合作以将语音输入快速地解码成单词。 第一层涉及声学数据，具有语音电话数据的第二层和第三层涉及字数据和字序列。 通过分离这些层，较高的层可以是时间独立的和异步的。 因此，异步层可以快速处理数据，并为第一层提供快速支持，从而将动态记录称为最可能连续语音结果的动态网络。 与先前已知的连续语音解码器相比，该解码器的速度和分离器允许更好的存储器效率和更好的解码器结果。

公开(公告)号：US5806029A

公开(公告)日：1998-09-08

申请号：US528821

申请日：1995-09-15

申请人： Eric Rolfe Buhrke ,  Wu Chou ,  Mazin G. Rahim

发明人： Eric Rolfe Buhrke ,  Wu Chou ,  Mazin G. Rahim

IPC分类号： G10L15/02 ,  G10L15/14 ,  G10L15/20 ,  G10L5/00

CPC分类号： G10L15/144 ,  G10L15/02 ,  G10L21/0272

摘要： Hierarchical signal bias removal (HSBR) signal conditioning uses a codebook constructed from the set of recognition models and is updated as the recognition models are modified during recognition model training. As a result, HSBR signal conditioning and recognition model training are based on the same set of recognition model parameters, which provides significant reduction in recognition error rate for the speech recognition system.

摘要翻译： 分级信号偏移去除（HSBR）信号调理使用由该组识别模型构建的码本，并且在识别模型训练期间识别模型被修改时被更新。 因此，HSBR信号调理和识别模型训练基于相同的识别模型参数集，这显着降低了语音识别系统的识别误码率。

公开(公告)号：US5797123A

公开(公告)日：1998-08-18

申请号：US771732

申请日：1996-12-20

申请人： Wu Chou ,  Biing-Hwang Juang ,  Tatsuya Kawahara ,  Chin-Hui Lee

发明人： Wu Chou ,  Biing-Hwang Juang ,  Tatsuya Kawahara ,  Chin-Hui Lee

IPC分类号： G10L15/10 ,  G10L15/00 ,  G10L15/08 ,  G10L15/18 ,  G10L15/28 ,  G10L9/00

CPC分类号： G10L15/18 ,  G10L15/1815 ,  G10L2015/088

摘要： A key-phrase detection and verification method that can be advantageously used to realize understanding of flexible (i.e., unconstrained) speech. A "multiple pass" procedure is applied to a spoken utterance comprising a sequence of words (i.e., a "sentence"). First, a plurality of key-phrases are detected (i.e., recognized) based on a set of phrase sub-grammars which may, for example, be specific to the state of the dialogue. These key-phrases are then verified by assigning confidence measures thereto and comparing these confidence measures to a threshold, resulting in a set of verified key-phrase candidates. Next, the verified key-phrase candidates are connected into sentence hypotheses based upon the confidence measures and predetermined (e.g., task-specific) semantic information. And, finally, one or more of these sentence hypotheses are verified to produce a verified sentence hypothesis and, from that, a resultant understanding of the spoken utterance.

摘要翻译： 可以有利地用于实现对柔性（即，无约束）语音的理解的密钥短语检测和验证方法。 对包括一系列单词（即“句子”）的语音话语应用“多重通行证”程序。 首先，基于短语子语法的集合来检测（即，识别）多个关键短语，其可以例如特定于对话状态。 然后通过分配置信度来验证这些密钥短语，并将这些置信度量度与阈值进行比较，得到一组已验证的密钥短语候选。 接下来，基于置信度度量和预定（例如，任务特定）语义信息将经验证的关键词候选者连接成句子假设。 最后，验证这些句子假说中的一个或多个，以产生一个经过验证的句子假设，从而得到对口头发音的理解。

公开(公告)号：US5579436A

公开(公告)日：1996-11-26

申请号：US30895

申请日：1993-03-15

申请人： Wu Chou ,  Biing-Hwang Juang

发明人： Wu Chou ,  Biing-Hwang Juang

IPC分类号： G10L15/06 ,  G10L15/14 ,  G10L5/04 ,  G10L5/06

CPC分类号： G10L15/063 ,  G10L15/144

摘要： A system pattern-based speech recognition, e.g., a hidden Markov model (HMM) based speech recognizer using Viterbi scoring. The principle of minimum recognition error rate is applied by the present invention using discriminative training. Various issues related to the special structure of HMMs are presented. Parameter update expressions for HMMs are provided.

摘要翻译： 基于系统模式的语音识别，例如使用维特比计分的基于隐马尔可夫模型（HMM）的语音识别器。 本发明使用区分性训练来应用最小识别错误率的原理。 介绍了与HMM特殊结构有关的各种问题。 提供HMM的参数更新表达式。

公开(公告)号：US08782773B2

公开(公告)日：2014-07-15

申请号：US12570848

申请日：2009-09-30

申请人： Wu Chou ,  Lookman Yasin Fazal ,  Weiping Guo ,  Feng Liu ,  Zhi Qiang Zhao

发明人： Wu Chou ,  Lookman Yasin Fazal ,  Weiping Guo ,  Feng Liu ,  Zhi Qiang Zhao

IPC分类号： G06F21/00

CPC分类号： G06F9/54

摘要： A system for enabling communication between a first domain and a second domain is disclosed. At least the first domain is protected by a firewall. A first data-processing system is provided in the first domain and a second data-processing system provided in second domain. The second domain hosts an application that the first domain desires to access. To enable the communication between the two domains a tunnel is established through the firewall. The tunnel runs from the first data-processing system to the second data-processing system. The second data-processing system provides a web-proxy interface to interface to the application and also acts as a tunnel gateway.

摘要翻译： 公开了一种用于实现第一域和第二域之间的通信的系统。 至少第一个域由防火墙保护。 在第一域中提供第一数据处理系统和在第二域中提供的第二数据处理系统。 第二个域托管第一个域想要访问的应用程序。 为了实现两个域之间的通信，通过防火墙建立隧道。 隧道从第一个数据处理系统运行到第二个数据处理系统。 第二个数据处理系统提供一个web应用程序接口，用于与应用程序的接口，并且还充当隧道网关。

公开(公告)号：US08656388B2

公开(公告)日：2014-02-18

申请号：US13108424

申请日：2011-05-16

申请人： Wu Chou ,  Weiping Guo ,  Feng Liu ,  Zhi Qiang Zhao

发明人： Wu Chou ,  Weiping Guo ,  Feng Liu ,  Zhi Qiang Zhao

IPC分类号： G06F9/455

CPC分类号： G06F11/2097 ,  G06F11/1484 ,  G06F11/203 ,  G06F11/2038

摘要： High availability (HA) protection is provided for an executing virtual machine. At a checkpoint in the HA process, the active server suspends the virtual machine; and the active server copies dirty memory pages to a buffer. During the suspension of the virtual machine on the active host server, dirty memory pages are copied to a ring buffer. A copy process copies the dirty pages to a first location in the buffer. At a predetermined benchmark or threshold, a transmission process can begin. The transmission process can read data out of the buffer at a second location to send to the standby host. Both the copy and transmission processes can operate asynchronously on the ring buffer. The ring buffer cannot overflow because the transmission process continues to empty the ring buffer as the copy process continues. This arrangement allows for using smaller buffers and prevents buffer overflows, and thereby, it reduces the VM suspension time and improves the system efficiency.

摘要翻译： 为执行的虚拟机提供高可用性（HA）保护。 在HA进程的检查点，活动服务器挂起虚拟机; 并且活动服务器将脏内存页复制到缓冲区。 在活动主机服务器上暂停虚拟机时，脏内存页被复制到环形缓冲区。 复制过程将脏页复制到缓冲区中的第一个位置。 在预定的基准或阈值处，可以开始传输过程。 传输过程可以在第二位置从缓冲器读取数据以发送到备用主机。 复制和传输过程都可以在环形缓冲区上异步运行。 环形缓冲区不能溢出，因为传输过程在复制过程继续时继续清空环形缓冲区。 这种安排允许使用较小的缓冲区并防止缓冲区溢出，从而减少VM挂起时间并提高系统效率。

公开(公告)号：US08413145B2

公开(公告)日：2013-04-02

申请号：US12895230

申请日：2010-09-30

申请人： Wu Chou ,  Weiping Guo ,  Feng Liu ,  Zhi Qiang Zhao

发明人： Wu Chou ,  Weiping Guo ,  Feng Liu ,  Zhi Qiang Zhao

IPC分类号： G06F9/455

CPC分类号： G06F12/08 ,  G06F9/45533 ,  G06F11/1484 ,  G06F11/2038 ,  G06F11/2048 ,  G06F11/2097 ,  G06F2212/151

摘要： High availability (HA) protection is provided for an executing virtual machine. At a checkpoint in the HA process, the active server suspends the virtual machine; and the active server copies dirty memory pages to a buffer. During the suspension of the virtual machine on the active host server, dirty memory pages are copied to a ring buffer. A copy process copies the dirty pages to a first location in the buffer. At a predetermined benchmark or threshold, a transmission process can begin. The transmission process can read data out of the buffer at a second location to send to the standby host. Both the copy and transmission processes can operate substantially simultaneously on the ring buffer. As such, the ring buffer cannot overflow because the transmission process continues to empty the ring buffer as the copy process continues. This arrangement allows for smaller buffers and prevents buffer overflows.

摘要翻译： 为执行的虚拟机提供高可用性（HA）保护。 在HA进程的检查点，活动服务器挂起虚拟机; 并且活动服务器将脏内存页复制到缓冲区。 在活动主机服务器上暂停虚拟机时，脏内存页被复制到环形缓冲区。 复制过程将脏页复制到缓冲区中的第一个位置。 在预定的基准或阈值处，可以开始传输过程。 传输过程可以在第二位置从缓冲器读取数据以发送到备用主机。 复制和传输过程都可以在环形缓冲器上基本同时运行。 因此，环形缓冲区不能溢出，因为复制过程继续，传输过程继续清空环形缓冲区。 这种安排允许更小的缓冲区并防止缓冲区溢出。

公开(公告)号：US5805772A

公开(公告)日：1998-09-08

申请号：US366843

申请日：1994-12-30

申请人： Wu Chou ,  Biing-Hwang Juang ,  Chin-Hui Lee ,  Tatsuo Matsuoka

发明人： Wu Chou ,  Biing-Hwang Juang ,  Chin-Hui Lee ,  Tatsuo Matsuoka

IPC分类号： G10L15/02 ,  G10L15/08 ,  G10L15/18 ,  G10L9/00

CPC分类号： G10L15/08 ,  G10L15/197

摘要： Disclosed are systems, methods and articles of manufacture for performing high resolution N-best string hypothesization during speech recognition. A received input signal, representing a speech utterance, is processed utilizing a plurality of recognition models to generate one or more string hypotheses of the received input signal. The plurality of recognition models preferably include one or more inter-word context dependent models and one or more language models. A forward partial path map is produced according to the allophonic specifications of at least one of the inter-word context dependent models and the language models. The forward partial path map is traversed in the backward direction as a function of the allophonic specifications to generate the one or more string hypotheses. One or more of the recognition models may represent one phone words.

摘要翻译： 公开了用于在语音识别期间执行高分辨率N最佳字符串假设的系统，方法和制品。 使用多个识别模型来处理表示语音话音的接收到的输入信号，以产生所接收的输入信号的一个或多个字符串假设。 多个识别模型优选地包括一个或多个词间上下文相关模型和一个或多个语言模型。 根据至少一个字词上下文相关模型和语言模型的等式规范来产生前向部分路径图。 作为生成一个或多个字符串假设的等式规范的函数，向前方向遍历前向部分路径图。 一个或多个识别模型可以表示一个电话词。

公开(公告)号：US5737489A

公开(公告)日：1998-04-07

申请号：US528902

申请日：1995-09-15

申请人： Wu Chou ,  Biing-Hwang Juang ,  Chin-Hui Lee ,  Mazin G. Rahim

发明人： Wu Chou ,  Biing-Hwang Juang ,  Chin-Hui Lee ,  Mazin G. Rahim

IPC分类号： G09B19/06 ,  G10L15/06 ,  G10L15/10 ,  G10L15/14 ,  G10L15/22 ,  G10L15/28 ,  G10L5/06

CPC分类号： G10L15/144 ,  G10L15/063 ,  G10L15/10

摘要： In a speech recognition system, a recognition processor receives an unknown utterance signal as input. The recognition processor in response to the unknown utterance signal input accesses a recognition database and scores the utterance signal against recognition models in the recognition database to classify the unknown utterance and to generate a hypothesis speech signal. A verification processor receives the hypothesis speech signal as input to be verified. The verification processor accesses a verification database to test the hypothesis speech signal against verification models reflecting a preselected type of training stored in the verification database. Based on the verification test, the verification processor generates a confidence measure signal. The confidence measure signal can be compared against a verification threshold to determine the accuracy of the recognition decision made by the recognition processor.

摘要翻译： 在语音识别系统中，识别处理器接收未知的话音信号作为输入。 响应于未知话语信号输入的识别处理器访问识别数据库，并根据识别数据库中的识别模型对话音信号进行评分，以对未知话语进行分类并生成假设语音信号。 验证处理器接收假设语音信号作为待验证的输入。 验证处理器访问验证数据库以针对反映存储在验证数据库中的预选类型的训练的验证模型来测试假设语音信号。 基于验证测试，验证处理器产生置信度测量信号。 可以将置信度信号与验证阈值进行比较，以确定由识别处理器进行的识别决策的准确性。