利索-全球专利检索

  1. 登录
  2. 注册

搜索结果

30 条记录 (耗时 0.022 秒)

    Systems and methods that detect a desired signal via a linear discriminative classifier that utilizes an estimated posterior signal-to-noise ratio (SNR)
    1.
    发明授权
    Systems and methods that detect a desired signal via a linear discriminative classifier that utilizes an estimated posterior signal-to-noise ratio (SNR) 有权
    通过利用估计的后验信噪比（SNR）的线性鉴别分类器来检测期望信号的系统和方法，

    公开(公告)号：US07660713B2

    公开(公告)日：2010-02-09

    申请号：US10795618

    申请日：2004-03-08

    申请人： Arungunram C. Surendran ,  John C. Platt ,  Somsak Sukittanon ,  Christopher J. C. Burges

    发明人： Arungunram C. Surendran ,  John C. Platt ,  Somsak Sukittanon ,  Christopher J. C. Burges

    IPC分类号： G10L21/00

    CPC分类号： G06K9/00536 ,  G10L25/78

    摘要： The present invention provides systems and methods for signal detection and enhancement. The systems and methods utilize one or more discriminative classifiers (e.g., a logistic regression model and a convolutional neural network) to estimate a posterior probability that indicates whether a desired signal is present in a received signal. The discriminative estimators generate the estimated probability based on one or more signal-to-noise ratio (SNRs) (e.g., a normalized logarithmic posterior SNR (nlpSNR) and a mel-transformed nlpSNR (mel-nlpSNR)) and an estimated noise model. Depending on the resolution desired, the estimated SNR can be generated at a frame level or at an atom level, wherein the atom level estimates are utilized to generate the frame level estimate. The novel systems and methods can be utilized to facilitate speech detection, speech recognition, speech coding, noise adaptation, speech enhancement, microphone arrays and echo-cancellation.

    摘要翻译： 本发明提供了用于信号检测和增强的系统和方法。 系统和方法利用一个或多个鉴别分类器（例如，逻辑回归模型和卷积神经网络）来估计指示所接收信号中是否存在期望信号的后验概率。 鉴别估计器基于一个或多个信噪比（SNR）（例如，归一化对数后验SNR（nlpSNR）和mel变换的nlpSNR（mel-nlpSNR））和估计的噪声模型来生成估计概率。 根据期望的分辨率，估计的SNR可以在帧级或原子级产生，其中原子级估计用于生成帧级估计。 可以利用新颖的系统和方法来促进语音检测，语音识别，语音编码，噪声适应，语音增强，麦克风阵列和回声消除。

    Leveraging unlabeled data with a probabilistic graphical model
    3.
    发明授权
    Leveraging unlabeled data with a probabilistic graphical model 有权
    利用概率图形模型利用未标记的数据

    公开(公告)号：US07937264B2

    公开(公告)日：2011-05-03

    申请号：US11170989

    申请日：2005-06-30

    申请人： Christopher J. C. Burges ,  John C. Platt

    发明人： Christopher J. C. Burges ,  John C. Platt

    IPC分类号： G06F17/27

    CPC分类号： G06F17/3071

    摘要： A general probabilistic formulation referred to as ‘Conditional Harmonic Mixing’ is provided, in which links between classification nodes are directed, a conditional probability matrix is associated with each link, and where the numbers of classes can vary from node to node. A posterior class probability at each node is updated by minimizing a divergence between its distribution and that predicted by its neighbors. For arbitrary graphs, as long as each unlabeled point is reachable from at least one training point, a solution generally always exists, is unique, and can be found by solving a sparse linear system iteratively. In one aspect, an automated data classification system is provided. The system includes a data set having at least one labeled category node in the data set. A semi-supervised learning component employs directed arcs to determine the label of at least one other unlabeled category node in the data set.

    摘要翻译： 提供了称为“条件谐波混合”的一般概率公式，其中分类节点之间的链接被引导，条件概率矩阵与每个链路相关联，并且类的数量可以在节点之间变化。 通过最小化其分布与其邻居预测的分布之间的差异来更新每个节点处的后级概率。 对于任意图，只要每个未标记的点从至少一个训练点到达，则通常总是存在的解是唯一的，并且可以通过迭代地求解稀疏线性系统来找到。 一方面，提供了一种自动数据分类系统。 该系统包括在数据集中具有至少一个标记类别节点的数据集。 半监督学习组件使用有向弧来确定数据集中至少一个其他未标记类别节点的标签。

    Probability estimate for K-nearest neighbor
    6.
    发明授权
    Probability estimate for K-nearest neighbor 有权
    K最近邻的概率估计

    公开(公告)号：US07016884B2

    公开(公告)日：2006-03-21

    申请号：US10183213

    申请日：2002-06-27

    申请人： John C. Platt ,  Christopher J. C. Burges

    发明人： John C. Platt ,  Christopher J. C. Burges

    IPC分类号： G06N3/02

    CPC分类号： G06K9/6276

    摘要： Systems and methods are disclosed that facilitate producing probabilistic outputs also referred to as posterior probabilities. The probabilistic outputs include an estimate of classification strength. The present invention intercepts non-probabilistic classifier output and applies a set of kernel models based on a softmax function to derive the desired probabilistic outputs. Such probabilistic outputs can be employed with handwriting recognition where the probability of a handwriting sample classification is combined with language models to make better classification decisions.

    摘要翻译： 公开了有助于产生也称为后验概率的概率输出的系统和方法。 概率输出包括分类强度的估计。 本发明拦截非概率分类器输出并且基于softmax函数应用一组核心模型以导出所需的概率输出。 这样的概率输出可以与手写识别一起使用，其中手写样本分类的概率与语言模型组合以进行更好的分类决定。

    Probability estimate for K-nearest neighbor
    7.
    发明授权
    Probability estimate for K-nearest neighbor 有权
    K最近邻的概率估计

    公开(公告)号：US07451123B2

    公开(公告)日：2008-11-11

    申请号：US11296919

    申请日：2005-12-08

    申请人： John C. Platt ,  Christopher J. C. Burges

    发明人： John C. Platt ,  Christopher J. C. Burges

    IPC分类号： G06E1/00 ,  G06F15/00

    CPC分类号： G06K9/6276

    摘要： Systems and methods are disclosed that facilitate producing probabilistic outputs also referred to as posterior probabilities. The probabilistic outputs include an estimate of classification strength. The present invention intercepts non-probabilistic classifier output and applies a set of kernel models based on a softmax function to derive the desired probabilistic outputs. Such probabilistic outputs can be employed with handwriting recognition where the probability of a handwriting sample classification is combined with language models to make better classification decisions.

    摘要翻译： 公开了有助于产生也称为后验概率的概率输出的系统和方法。 概率输出包括分类强度的估计。 本发明拦截非概率分类器输出并且基于softmax函数应用一组核心模型以导出所需的概率输出。 这样的概率输出可以与手写识别一起使用，其中手写样本分类的概率与语言模型组合以进行更好的分类决定。

    Audio duplicate detector
    9.
    发明授权
    Audio duplicate detector 有权
    音频重复检测器

    公开(公告)号：US07421305B2

    公开(公告)日：2008-09-02

    申请号：US10785561

    申请日：2004-02-24

    申请人： Christopher J. C. Burges ,  John C. Platt ,  Daniel Plastina ,  Erin L. Renshaw

    发明人： Christopher J. C. Burges ,  John C. Platt ,  Daniel Plastina ,  Erin L. Renshaw

    IPC分类号： G06F17/00

    CPC分类号： G06F17/30758 ,  G06F17/30017 ,  G06F17/30743 ,  G10H1/0041 ,  G10H2240/061 ,  G10H2240/155 ,  G10H2250/221 ,  G11B27/28

    摘要： The present invention relates to a system and methodology to facilitate automatic management and pruning of audio files residing in a database. Audio fingerprinting is a powerful tool for identifying streaming or file-based audio, using a database of fingerprints. Duplicate detection identifies duplicate audio clips in a set, even if the clips differ in compression quality or duration. The present invention can be provided as a self-contained application that it does not require an external database of fingerprints. Also, a user interface provides various options for managing and pruning the audio files.

    摘要翻译： 本发明涉及一种便于自动管理和修剪驻留在数据库中的音频文件的系统和方法。 音频指纹是使用指纹数据库识别流媒体或基于文件的音频的强大工具。 重复的检测识别集合中的重复音频剪辑，即使剪辑在压缩质量或持续时间上有所不同。 本发明可以作为独立应用来提供，其不需要外部指纹数据库。 此外，用户界面提供了管理和修剪音频文件的各种选项。

    Identification of sample data items for re-judging
    10.
    发明授权
    Identification of sample data items for re-judging 有权
    识别重新评估的样本数据项

    公开(公告)号：US08935258B2

    公开(公告)日：2015-01-13

    申请号：US12484256

    申请日：2009-06-15

    申请人： Krysta M. Svore ,  Elbio Renato Torres Abib ,  Christopher J. C. Burges ,  Bhuvan Middha

    发明人： Krysta M. Svore ,  Elbio Renato Torres Abib ,  Christopher J. C. Burges ,  Bhuvan Middha

    IPC分类号： G06F17/30 ,  G06K9/62

    CPC分类号： G06F17/30864 ,  G06K9/6257

    摘要： Described is a technology for identifying sample data items (e.g., documents corresponding to query-URL pairs) having the greatest likelihood of being mislabeled when previously judged, and selecting those data items for re-judging. In one aspect, lambda gradient scores (information associated with ranked sample data items that indicates a relative direction and how “strongly” to move each data item for lowering a ranking cost) are summed for pairs of sample data items to compute re-judgment scores for each of those sample data items. The re-judgment scores indicate a relative likelihood of mislabeling. Once the selected sample data items are re-judged, a new training set is available, whereby a new ranker may be trained.

    摘要翻译： 描述了用于在先前判断时识别具有错误标记的最大可能性的样本数据项（例如，与查询 - URL对相对应的文档）的技术，并且选择那些用于重新判断的数据项。 在一个方面，λ梯度分数（与指示相对方向的排名样本数据项相关联的信息以及如何“强烈地”移动每个数据项以降低排名成本）被求和于样本数据项对以计算重新判断分数 对于每个样本数据项。 重新判断分数表明错误标签的相对可能性。 一旦选择的样本数据项被重新判断，就可以使用新的训练集，由此可以训练新的训练者。