Covariance estimation for pattern recognition
    1.
    发明申请
    Covariance estimation for pattern recognition 有权
    模式识别的协方差估计

    公开(公告)号:US20070005355A1

    公开(公告)日:2007-01-04

    申请号:US11173907

    申请日:2005-07-01

    IPC分类号: G10L15/00

    CPC分类号: G06K9/6297 G10L15/02

    摘要: A reliable full covariance matrix estimation algorithm for pattern unit's state output distribution in pattern recognition system is discussed. An intermediate hierarchical tree structure is built to relate models for product units. Full covariance matrices of pattern unit's state output distribution are estimated based on all the related nodes in the tree.

    摘要翻译: 讨论了模式识别系统中模式单元状态输出分布的可靠全协方差矩阵估计算法。 建立了一个中间分层树结构来关联产品单元的模型。 基于树中所有相关节点估计模式单位状态输出分布的全协方差矩阵。

    Covariance estimation for pattern recognition
    3.
    发明授权
    Covariance estimation for pattern recognition 有权
    模式识别的协方差估计

    公开(公告)号:US07805301B2

    公开(公告)日:2010-09-28

    申请号:US11173907

    申请日:2005-07-01

    IPC分类号: G10L15/14

    CPC分类号: G06K9/6297 G10L15/02

    摘要: A reliable full covariance matrix estimation algorithm for pattern unit's state output distribution in pattern recognition system is discussed. An intermediate hierarchical tree structure is built to relate models for product units. Full covariance matrices of pattern unit's state output distribution are estimated based on all the related nodes in the tree.

    摘要翻译: 讨论了模式识别系统中模式单元状态输出分布的可靠全协方差矩阵估计算法。 建立了一个中间分层树结构来关联产品单元的模型。 基于树中所有相关节点估计模式单位状态输出分布的全协方差矩阵。

    Subword unit posterior probability for measuring confidence
    5.
    发明授权
    Subword unit posterior probability for measuring confidence 有权
    子字单位后验概率用于测量置信度

    公开(公告)号:US07890325B2

    公开(公告)日:2011-02-15

    申请号:US11376803

    申请日:2006-03-16

    IPC分类号: G06F17/27 G10L15/00 G10L15/28

    摘要: Speech recognition such as command and control speech recognition generally use a context free grammar to constrain the decoding process. Word or subword background model are constructed to repopulate dynamic hypothesis space, especially when word spareness is at issue. The background models can be later used in speech recognition. During speech recognition, background and conventional context free grammar decoding are used to measure confidence. The discussion above is merely provided for general background information and is not intended to be used as an aid in determining the scope of the claimed subject matter.

    摘要翻译: 诸如命令和控制语音识别之类的语音识别通常使用无上下文的语法来限制解码过程。 构建词或子词背景模型,以重新构建动态假设空间,特别是在词语空间问题时。 背景模型可以稍后用于语音识别。 在语音识别期间,使用背景和常规上下文无关语法解码来测量置信度。 上面的讨论仅用于一般背景信息,并不旨在用于帮助确定所要求保护的主题的范围。

    Subword unit posterior probability for measuring confidence
    6.
    发明申请
    Subword unit posterior probability for measuring confidence 有权
    子字单位后验概率用于测量置信度

    公开(公告)号:US20070219797A1

    公开(公告)日:2007-09-20

    申请号:US11376803

    申请日:2006-03-16

    IPC分类号: G10L15/18

    摘要: Speech recognition such as command and control speech recognition generally use a context free grammar to constrain the decoding process. Word or subword background model are constructed to repopulate dynamic hypothesis space, especially when word spareness is at issue. The background models can be later used in speech recognition. During speech recognition, background and conventional context free grammar decoding are used to measure confidence. The discussion above is merely provided for general background information and is not intended to be used as an aid in determining the scope of the claimed subject matter.

    摘要翻译: 诸如命令和控制语音识别之类的语音识别通常使用无上下文的语法来限制解码过程。 构建词或子词背景模型,以重新构建动态假设空间,特别是在词语空间问题时。 背景模型可以稍后用于语音识别。 在语音识别期间,使用背景和常规上下文无关语法解码来测量置信度。 上面的讨论仅用于一般背景信息,并不旨在用于帮助确定所要求保护的主题的范围。

    Calculating cost measures between HMM acoustic models
    7.
    发明申请
    Calculating cost measures between HMM acoustic models 有权
    计算HMM声学模型之间的成本测量

    公开(公告)号:US20080059184A1

    公开(公告)日:2008-03-06

    申请号:US11507859

    申请日:2006-08-22

    IPC分类号: G10L15/14

    CPC分类号: G10L15/142

    摘要: Measurement of Kullback-Leibler Divergence (KLD) between hidden Markov models (HMM) of acoustic units utilizes an unscented transform to approximate KLD between Gaussian mixtures. Dynamic programming equalizes the number of states between HMMs having a different number of states, while the total KLD of the HMMs is obtained by summing individual KLDs calculated by state pair by state pair comparisons.

    摘要翻译: 声学单元的隐马尔可夫模型(HMM)之间的Kullback-Leibler发散(KLD)的测量利用无差异变换来近似高斯混合之间的KLD。 动态规划使具有不同数量状态的HMM之间的状态数量相等,而HMM的总KLD是通过将通过状态对比较的状态对计算的各个KLD求和来获得的。

    Common word graph based multimodal input
    8.
    发明申请
    Common word graph based multimodal input 有权
    基于常用字图的多模态输入

    公开(公告)号:US20070239432A1

    公开(公告)日:2007-10-11

    申请号:US11394809

    申请日:2006-03-30

    IPC分类号: G06F17/27

    CPC分类号: G06F17/27

    摘要: Multiple input modalities are selectively used by a user or process to prune a word graph. Pruning initiates rescoring in order to generate a new word graph with a revised best path.

    摘要翻译: 用户或进程有选择地使用多种输入模式来修剪单词图形。 修剪开始拯救,以生成一个修改最佳路径的新字图。

    Method and apparatus for tracking pitch in audio analysis
    9.
    发明授权
    Method and apparatus for tracking pitch in audio analysis 失效
    音频分析跟踪音调的方法和装置

    公开(公告)号:US06917912B2

    公开(公告)日:2005-07-12

    申请号:US09843212

    申请日:2001-04-24

    IPC分类号: G10L25/90 G10L11/04

    CPC分类号: G10L25/90

    摘要: A computationally efficient and robust pitch detection and tracking system and related methods are presented. According to certain exemplary implementations a method is presented comprising identifying an initial set of pitch period candidates using a first estimation algorithm, filtering the initial set of candidates and passing the filtered candidates through a second, more accurate pitch estimation algorithm to generate a final set of pitch period candidates from which the most likely pitch value is selected.

    摘要翻译: 提出了一种计算有效和鲁棒的音高检测和跟踪系统及相关方法。 根据某些示例性实施方式,呈现一种方法,包括使用第一估计算法来识别初始的音调周期候选集合,对候选的初始集合进行滤波,并且通过第二更精确的音调估计算法传递经滤波的候选,以产生最终的一组 选择最可能的音调值的音调周期候选。