公开(公告)号：US20090006094A1

公开(公告)日：2009-01-01

申请号：US12206384

申请日：2008-09-08

申请人： Jiri Navratil ,  Ganesh N. Ramaswamy

发明人： Jiri Navratil ,  Ganesh N. Ramaswamy

IPC分类号： G10L17/00

CPC分类号： G10L17/06

摘要： In detection systems, such as speaker verification systems, for a given operating point range, with an associated detection “cost”, the detection cost is preferably reduced by essentially trading off the system error in the area of interest with areas essentially “outside” that interest. Among the advantages achieved thereby are higher optimization gain and better generalization. From a measurable Detection Error Tradeoff (DET) curve of the given detection system, a criterion is preferably derived, such that its minimization provably leads to detection cost reduction in the area of interest. The criterion allows for selective access to the slope and offset of the DET curve (a line in case of normally distributed detection scores, a curve approximated by mixture of Gaussians in case of other distributions). By modifying the slope of the DET curve, the behavior of the detection system is changed favorably with respect to the given area of interest.

摘要翻译： 在诸如扬声器验证系统的检测系统中，对于给定的操作点范围，具有相关联的检测“成本”，优选地通过基本上“感兴趣的区域”内的基本上“外部”区域的系统误差基本上消除了检测成本， 利益。 其中实现的优点是优化增益更高，更好的泛化。 从给定检测系统的可测量的检测误差折衷（DET）曲线中，优选地导出标准，使得其最小化可以导致感兴趣区域中的检测成本降低。 该标准允许选择性地访问DET曲线的斜率和偏移（在正态分布的检测分数的情况下的线，在其他分布的情况下高斯混合的曲线近似的曲线）。 通过修改DET曲线的斜率，检测系统的行为相对于给定的感兴趣区域有利地改变。

公开(公告)号：US20080257047A1

公开(公告)日：2008-10-23

申请号：US12128321

申请日：2008-05-28

申请人： Jason William Pelecanos ,  Douglas George Heintzman ,  Jiri Navratil ,  Ganesh N. Ramaswamy

发明人： Jason William Pelecanos ,  Douglas George Heintzman ,  Jiri Navratil ,  Ganesh N. Ramaswamy

IPC分类号： G01H17/00 ,  G06F19/00

CPC分类号： G01H1/00

摘要： A method, system and program storage device are provided for machine diagnostics, detection and profiling using pressure waves, the method including profiling known sources, acquiring pressure wave data, analyzing the acquired pressure wave data, and detecting if the analyzed pressure wave data matches a profiled known source; the system including a processor, a pressure wave transducer in signal communication with the processor, a pressure wave analysis unit in signal communication with the processor, and a source or threat detection unit in signal communication with the processor; and the program storage device including program steps for profiling known sources, acquiring pressure wave data, analyzing the acquired pressure wave data, and detecting if the analyzed pressure wave data matches a profiled known source.

摘要翻译： 提供了一种方法，系统和程序存储装置，用于使用压力波的机器诊断，检测和轮廓，该方法包括对已知源进行分析，获取压力波数据，分析获取的压力波数据，以及检测分析的压力波数据是否匹配 剖析已知来源; 所述系统包括处理器，与所述处理器进行信号通信的压力波换能器，与所述处理器进行信号通信的压力波分析单元，以及与所述处理器进行信号通信的源或威胁检测单元; 程序存储装置包括用于对已知源进行分析的程序步骤，获取压力波数据，分析获取的压力波数据，以及检测所分析的压力波数据是否与分析的已知来源相匹配。

公开(公告)号：US20080222722A1

公开(公告)日：2008-09-11

申请号：US12057470

申请日：2008-03-28

申请人： Jiri Navratil ,  Ryan L. Osborn ,  Jason W. Pelecanos ,  Ganesh N. Ramaswamy ,  Ran D. Zilca

发明人： Jiri Navratil ,  Ryan L. Osborn ,  Jason W. Pelecanos ,  Ganesh N. Ramaswamy ,  Ran D. Zilca

IPC分类号： H04L9/32

CPC分类号： H04L63/0861 ,  G06F21/31 ,  G06F2221/2103 ,  H04L63/104

摘要： Methods and apparatus are provided for sequential authentication of a user that employ one or mole error rates characterizing each security challenge. According to one aspect of the invention, a user is challenged with at least one knowledge challenge to obtain an intermediate authentication result; and the user challenges continue until a cumulative authentication result satisfies one or more criteria. The intermediate authentication result is based, for example, on one or more of false accept and false reject error probabilities for each knowledge challenge. A false accept error probability describes a probability of a different user answering the knowledge challenge correctly. A false reject error probability describes a probability of a genuine user not answering the knowledge challenge correctly. The false accept and false reject error probabilities can be adapted based on field data or known information about a given challenge.

摘要翻译： 提供了用于对使用表征每个安全挑战的一个或多个误差率的用户的顺序验证的方法和装置。 根据本发明的一个方面，用户被挑战至少一个知识挑战以获得中间认证结果; 并且用户挑战持续，直到累积认证结果满足一个或多个标准。 中间认证结果例如基于每个知识挑战的一个或多个假接收和错误拒绝错误概率。 错误接受误差概率描述了不同用户正确回答知识挑战的概率。 错误拒绝错误概率描述了正确用户没有正确回答知识挑战的概率。 可以基于现场数据或关于给定挑战的已知信息来适应错误接受和错误拒绝错误概率。

公开(公告)号：US20080052072A1

公开(公告)日：2008-02-28

申请号：US11930970

申请日：2007-10-31

申请人： Upendra Chaudhari ,  Stephane Maes ,  Jiri Navratil

发明人： Upendra Chaudhari ,  Stephane Maes ,  Jiri Navratil

IPC分类号： G10L15/00

CPC分类号： G10L17/14 ,  G10L17/08

摘要： A technique for improved score calculation and normalization in a framework of recognition with phonetically structured speaker models. The technique involves determining, for each frame and each level of phonetic detail of a target speaker model, a non-interpolated likelihood value, and then resolving the at least one likelihood value to obtain a likelihood score.

摘要翻译： 一种在语音结构化的扬声器模型识别框架中改进分数计算和归一化的技术。 该技术涉及针对目标说话者模型的每个帧和每个级别的语音细节确定非内插似然值，然后解析该至少一个似然值以获得似然分数。

公开(公告)号：US20070239441A1

公开(公告)日：2007-10-11

申请号：US11391891

申请日：2006-03-29

申请人： Jiri Navratil ,  Jason Pelecanos ,  Ganesh Ramaswamy

发明人： Jiri Navratil ,  Jason Pelecanos ,  Ganesh Ramaswamy

IPC分类号： G10L19/14 ,  G10L21/00

CPC分类号： G10L17/20

摘要： A method and system for speaker recognition and identification includes transforming features of a speaker utterance in a first condition state to match a second condition state and provide a transformed utterance. A discriminative criterion is used to generate a transform that maps an utterance to obtain a computed result. The discriminative criterion is maximized over a plurality of speakers to obtain a best transform for recognizing speech and/or identifying a speaker under the second condition state. Speech recognition and speaker identity may be determined by employing the best transform for decoding speech to reduce channel mismatch.

摘要翻译： 用于说话人识别和识别的方法和系统包括在第一状况状态下转换说话人话语的特征以匹配第二状态，并提供转换后的话语。 使用歧视性标准来产生映射发音以获得计算结果的变换。 歧视性标准在多个扬声器上最大化，以获得用于在第二状态下识别语音和/或识别扬声器的最佳变换。 可以通过采用用于解码语音的最佳变换来减少信道不匹配来确定语音识别和扬声器身份。

公开(公告)号：US07162641B1

公开(公告)日：2007-01-09

申请号：US09592310

申请日：2000-06-13

申请人： Upendra V. Chaudhari ,  Stephane H. Maes ,  Jiri Navratil

发明人： Upendra V. Chaudhari ,  Stephane H. Maes ,  Jiri Navratil

IPC分类号： H04K1/00 ,  H04L9/00 ,  G10L15/00 ,  G10L17/00

CPC分类号： G10L17/06

摘要： Methods and apparatus for providing speech-based authentication, including the determination of a target discriminant based on an identity claim and on at least one target voiceprint model relating to a target speaker, of a background discriminant based on the identity claim and on at least one background voiceprint model relating to at least one background speaker, and of a score based on the target discriminant and the background discriminant, which score is used to accept or reject the identity claim.

摘要翻译： 用于基于身份声明提供基于语音的认证的方法和装置，包括基于身份声明的目标判别式的确定和与目标说话者有关的至少一个目标声纹模型基于身份声明的背景判别式以及至少一个 与至少一个背景说话者相关的背景声纹模型以及基于目标判别式和背景判别式的得分，该分数用于接受或拒绝身份声明。

公开(公告)号：US20060111905A1

公开(公告)日：2006-05-25

申请号：US10994743

申请日：2004-11-22

申请人： Jiri Navratil ,  James Nealand ,  Jason Pelecanos ,  Ganesh Ramaswamy ,  Ran Zilca

发明人： Jiri Navratil ,  James Nealand ,  Jason Pelecanos ,  Ganesh Ramaswamy ,  Ran Zilca

IPC分类号： G10L15/00

CPC分类号： G10L15/144 ,  G10L15/063

摘要： There is provided an apparatus for providing a Text Independent (TI) speaker recognition mode in a Text Dependent (TD) Hidden Markov Model (HMM) speaker recognition system and/or a Text Constrained (TC) HMM speaker recognition system. The apparatus includes a Gaussian Mixture Model (GMM) generator and a Gaussian weight normalizer. The GMM generator is for creating a GMM by pooling Gaussians from a plurality of HMM states. The Gaussian weight normalizer is for normalizing Gaussian weights with respect to the plurality of HMM states.

摘要翻译： 提供了一种用于在文本依赖（TD）隐马尔可夫模型（HMM）说话人识别系统和/或文本约束（TC）HMM说话者识别系统中提供文本独立（TI）说话者识别模式的装置。 该装置包括高斯混合模型（GMM）发生器和高斯权重归一化器。 GMM生成器用于通过从多个HMM状态汇集高斯来创建GMM。 高斯权重规范化器用于相对于多个HMM状态归一化高斯权重。

公开(公告)号：US08441777B2

公开(公告)日：2013-05-14

申请号：US12474309

申请日：2009-05-29

申请人： Stanislav Zednicek ,  Peter Jacisko ,  Tomas Karnik ,  Jiri Navratil

发明人： Stanislav Zednicek ,  Peter Jacisko ,  Tomas Karnik ,  Jiri Navratil

IPC分类号： H01G9/02 ,  H01G9/04

CPC分类号： H01G9/012 ,  H01G9/10 ,  H01G9/15 ,  Y10T29/417

摘要： A solid electrolytic capacitor that contains a capacitor element that includes an anode body, dielectric layer, and solid electrolyte is provided. The capacitor also contains an anode lead that is electrically connected to the anode body by a refractory metal paste (e.g., tantalum paste). The use of such a refractory metal paste allows the anode lead to be sinter bonded to a surface of the anode body after it is pressed. In this manner, a strong and reliable connection may be achieved without substantially decreasing the surface area of the lead that is available for connection to a termination. Furthermore, because the lead is not embedded within the anode body, the capacitor may be configured so that little, if any, portion of the lead extends beyond the anode body. This may result in a highly volumetrically efficient capacitor with excellent electrical properties.

摘要翻译： 提供一种含有包括阳极体，电介质层和固体电解质的电容器元件的固体电解电容器。 电容器还包含通过难熔金属膏（例如，钽浆）与阳极体电连接的阳极引线。 使用这样的难熔金属糊剂可以使阳极引线在压制之后烧结到阳极体的表面。 以这种方式，可以实现强而可靠的连接，而不会显着减少可用于连接到终端的引线的表面积。 此外，由于引线未嵌入在阳极体内，所以电容器可以被配置成使得很少（如果有的话）引线的部分延伸超出阳极体。 这可能导致具有优异电性能的高体积效率的电容器。

公开(公告)号：US20110191096A1

公开(公告)日：2011-08-04

申请号：US12697047

申请日：2010-01-29

申请人： Ruhi Sarikaya ,  Jiri Navratil ,  Osamuyimen Stewart ,  David Lubensky

发明人： Ruhi Sarikaya ,  Jiri Navratil ,  Osamuyimen Stewart ,  David Lubensky

IPC分类号： G06F17/28 ,  A63F9/24

CPC分类号： A63F9/24 ,  G06F17/28

摘要： A method of generating a statistical machine translation database through a game in which a monolingual structure is provided to a plurality of players. A first translation attempt is received from each of the plurality of players. The first translation attempt from each of the plurality of players is compared. Feedback is provided to each of the plurality of players and the attempts are received and compared to provide feedback to iteratively converge subsequent translations from each of the plurality of players into a final translated structure.

摘要翻译： 一种通过游戏产生统计机器翻译数据库的方法，其中向多个玩家提供单语构造。 从多个玩家中的每一个接收第一翻译尝试。 比较来自多个玩家中的每一个的第一翻译尝试。 向多个玩家中的每一个提供反馈，并且尝试被接收并且进行比较以提供反馈以迭代地收敛从多个玩家中的每一个到随后的翻译结构的后续翻译。

公开(公告)号：US07970614B2

公开(公告)日：2011-06-28

申请号：US11745745

申请日：2007-05-08

申请人： Janice J. Kim ,  Jiri Navratil ,  Jason W. Pelecanos ,  Ganesh N. Ramaswamy

发明人： Janice J. Kim ,  Jiri Navratil ,  Jason W. Pelecanos ,  Ganesh N. Ramaswamy

IPC分类号： G10L15/14

CPC分类号： G10L15/20

摘要： The present invention provides a system and method for treating distortion propagated though a detection system. The system includes a compensation module that compensates for untreated distortions propagating through the detection compensation system, a user model pool that comprises of a plurality of model sets, and a model selector that selects at least one model set from plurality of model sets in the user model pool. The compensation is accomplished by continually producing scores distributed according to a prescribed distribution for the at least one model set and mitigating the adverse effects of the scores being distorted and lying off a pre-set operating point.The method for treating distortion propagated though a detection system includes receiving a signal from a remote device, and compensating the signal for untreated distortions. The compensation includes selecting at least one relevant model set from a plurality of model sets, producing scores distributed according to a pre-described distribution for the at least one model set, and mitigating the adverse effect of the scores being distorted by rejecting a signal if it lies off a preset operating point.

摘要翻译： 本发明提供一种用于处理通过检测系统传播的失真的系统和方法。 该系统包括补偿模块，其补偿通过检测补偿系统传播的未经处理的失真，包括多个模型组的用户模型池以及从用户中的多个模型集中选择至少一个模型集合的模型选择器 模型池。 补偿是通过不断产生根据至少一个模型集合的规定分布分配的分数来实现的，并且减轻被扭曲的分数和摆脱预设操作点的不利影响。 用于处理通过检测系统传播的失真的方法包括从远程设备接收信号，以及补偿信号以用于未经处理的失真。 所述补偿包括从多个模型集中选择至少一个相关模型集合，产生根据所述至少一个模型集合的预先描述的分布分配的分数，以及通过拒绝信号来减轻所述分数失真的不利影响，如果 它摆脱了预设的操作点。