Adapting a compressed model for use in speech recognition
    1.
    发明授权
    Adapting a compressed model for use in speech recognition 有权
    适应用于语音识别的压缩模型

    公开(公告)号:US08239195B2

    公开(公告)日:2012-08-07

    申请号:US12235748

    申请日:2008-09-23

    IPC分类号: G10L15/20

    CPC分类号: G10L15/20 G10L15/065

    摘要: A speech recognition system includes a receiver component that receives a distorted speech utterance. The speech recognition also includes an adaptor component that selectively adapts parameters of a compressed model used to recognize at least a portion of the distorted speech utterance, wherein the adaptor component selectively adapts the parameters of the compressed model based at least in part upon the received distorted speech utterance.

    摘要翻译: 语音识别系统包括接收失真的语音话语的接收机组件。 所述语音识别还包括适配器组件,所述适配器组件选择性地适配用于识别所述失真语音话语的至少一部分的压缩模型的参数,其中所述适配器组件至少部分地基于接收失真的语音话语选择性地调整所述压缩模型的参数 讲话话语。

    ADAPTING A COMPRESSED MODEL FOR USE IN SPEECH RECOGNITION
    2.
    发明申请
    ADAPTING A COMPRESSED MODEL FOR USE IN SPEECH RECOGNITION 有权
    适应用于语音识别的压缩模型

    公开(公告)号:US20100076757A1

    公开(公告)日:2010-03-25

    申请号:US12235748

    申请日:2008-09-23

    IPC分类号: G10L15/20

    CPC分类号: G10L15/20 G10L15/065

    摘要: A speech recognition system includes a receiver component that receives a distorted speech utterance. The speech recognition also includes an adaptor component that selectively adapts parameters of a compressed model used to recognize at least a portion of the distorted speech utterance, wherein the adaptor component selectively adapts the parameters of the compressed model based at least in part upon the received distorted speech utterance.

    摘要翻译: 语音识别系统包括接收失真的语音话语的接收机组件。 所述语音识别还包括适配器组件,所述适配器组件选择性地适配用于识别所述失真语音话语的至少一部分的压缩模型的参数,其中所述适配器组件至少部分地基于接收失真的语音话语选择性地调整所述压缩模型的参数 讲话话语。

    Noise Suppressor for Robust Speech Recognition
    3.
    发明申请
    Noise Suppressor for Robust Speech Recognition 有权
    噪声抑制器用于强大的语音识别

    公开(公告)号:US20100153104A1

    公开(公告)日:2010-06-17

    申请号:US12335558

    申请日:2008-12-16

    IPC分类号: G10L15/20

    CPC分类号: G10L21/0208 G10L15/20

    摘要: Described is noise reduction technology generally for speech input in which a noise-suppression related gain value for the frame is determined based upon a noise level associated with that frame in addition to the signal to noise ratios (SNRs). In one implementation, a noise reduction mechanism is based upon minimum mean square error, Mel-frequency cepstra noise reduction technology. A high gain value (e.g., one) is set to accomplish little or no noise suppression when the noise level is below a threshold low level, and a low gain value set or computed to accomplish large noise suppression above a threshold high noise level. A noise-power dependent function, e.g., a log-linear interpolation, is used to compute the gain between the thresholds. Smoothing may be performed by modifying the gain value based upon a prior frame's gain value. Also described is learning parameters used in noise reduction via a step-adaptive discriminative learning algorithm.

    摘要翻译: 描述了通常用于语音输入的噪声降低技术,其中除了信噪比(SNR)之外,基于与该帧相关联的噪声电平来确定用于帧的噪声抑制相关增益值。 在一个实现中,降噪机制基于最小均方误差,Mel-frequency cepstra降噪技术。 设置高增益值(例如一个),以在噪声电平低于阈值低电平时实现很少或没有噪声抑制,以及设置或计算的低增益值,以实现高于阈值高噪声电平的大噪声抑制。 使用噪声功率相关函数,例如对数线性插值来计算阈值之间的增益。 可以通过基于先前帧的增益值修改增益值来执行平滑化。 还描述了通过步进自适应识别学习算法在降噪中使用的学习参数。

    Noise suppressor for robust speech recognition
    4.
    发明授权
    Noise suppressor for robust speech recognition 有权
    噪声抑制器用于强大的语音识别

    公开(公告)号:US08185389B2

    公开(公告)日:2012-05-22

    申请号:US12335558

    申请日:2008-12-16

    IPC分类号: G10L15/20

    CPC分类号: G10L21/0208 G10L15/20

    摘要: Described is noise reduction technology generally for speech input in which a noise-suppression related gain value for the frame is determined based upon a noise level associated with that frame in addition to the signal to noise ratios (SNRs). In one implementation, a noise reduction mechanism is based upon minimum mean square error, Mel-frequency cepstra noise reduction technology. A high gain value (e.g., one) is set to accomplish little or no noise suppression when the noise level is below a threshold low level, and a low gain value set or computed to accomplish large noise suppression above a threshold high noise level. A noise-power dependent function, e.g., a log-linear interpolation, is used to compute the gain between the thresholds. Smoothing may be performed by modifying the gain value based upon a prior frame's gain value. Also described is learning parameters used in noise reduction via a step-adaptive discriminative learning algorithm.

    摘要翻译: 描述了通常用于语音输入的噪声降低技术,其中除了信噪比(SNR)之外,基于与该帧相关联的噪声电平来确定用于帧的噪声抑制相关增益值。 在一个实现中,降噪机制基于最小均方误差,Mel-frequency cepstra降噪技术。 设置高增益值(例如一个),以在噪声电平低于阈值低电平时实现很少或没有噪声抑制,以及设置或计算的低增益值,以实现高于阈值高噪声电平的大噪声抑制。 使用噪声功率相关函数,例如对数线性插值来计算阈值之间的增益。 可以通过基于先前帧的增益值修改增益值来执行平滑化。 还描述了通过步进自适应识别学习算法在降噪中使用的学习参数。

    Method and apparatus for constructing a speech filter using estimates of clean speech and noise
    5.
    发明授权
    Method and apparatus for constructing a speech filter using estimates of clean speech and noise 有权
    用于使用干净的语音和噪声的估计来构造语音滤波器的方法和装置

    公开(公告)号:US07725314B2

    公开(公告)日:2010-05-25

    申请号:US10780177

    申请日:2004-02-16

    CPC分类号: G10L21/0208

    摘要: A method and apparatus identify a clean speech signal from a noisy speech signal. To do this, a clean speech value and a noise value are estimated from the noisy speech signal. The clean speech value and the noise value are then used to define a gain on a filter. The noisy speech signal is applied to the filter to produce the clean speech signal. Under some embodiments, the noise value and the clean speech value are used in both the numerator and the denominator of the filter gain, with the numerator being guaranteed to be positive.

    摘要翻译: 方法和装置从噪声语音信号中识别干净的语音信号。 为此,从噪声语音信号估计干净的语音值和噪声值。 然后使用干净的语音值和噪声值来定义滤波器上的增益。 噪声语音信号被施加到滤波器以产生干净的语音信号。 在一些实施例中,噪声值和清洁语音值用于滤波器增益的分子和分母,分子保证为正。

    Method and apparatus for constructing a speech filter using estimates of clean speech and noise
    6.
    发明申请
    Method and apparatus for constructing a speech filter using estimates of clean speech and noise 有权
    用于使用干净的语音和噪声的估计来构造语音滤波器的方法和装置

    公开(公告)号:US20050182624A1

    公开(公告)日:2005-08-18

    申请号:US10780177

    申请日:2004-02-16

    IPC分类号: G10L21/00 G10L21/02

    CPC分类号: G10L21/0208

    摘要: A method and apparatus identify a clean speech signal from a noisy speech signal. To do this, a clean speech value and a noise value are estimated from the noisy speech signal. The clean speech value and the noise value are then used to define a gain on a filter. The noisy speech signal is applied to the filter to produce the clean speech signal. Under some embodiments, the noise value and the clean speech value are used in both the numerator and the denominator of the filter gain, with the numerator being guaranteed to be positive.

    摘要翻译: 方法和装置从噪声语音信号中识别干净的语音信号。 为此,从噪声语音信号估计干净的语音值和噪声值。 然后使用干净的语音值和噪声值来定义滤波器上的增益。 噪声语音信号被施加到滤波器以产生干净的语音信号。 在一些实施例中,噪声值和清洁语音值用于滤波器增益的分子和分母,分子保证为正。

    DNA SEQUENCING BY SYNTHESIS USING RAMAN AND INFRARED SPECTROSCOPY DETECTION
    7.
    发明申请
    DNA SEQUENCING BY SYNTHESIS USING RAMAN AND INFRARED SPECTROSCOPY DETECTION 有权
    通过使用拉曼和红外光谱检测合成的DNA序列

    公开(公告)号:US20150080232A1

    公开(公告)日:2015-03-19

    申请号:US14119846

    申请日:2012-05-23

    IPC分类号: C12Q1/68

    摘要: This invention provides nucleoside triphosphate analogues having the structure: wherein B is a base and is adenine, guanine, cytosine, uracil or thymine, wherein R″ is an OH or an H, and wherein R′ is azidomethyl, a hydrocarbyl, or a substituted hydrocarbyl, and which has a Raman spectroscopy peak with wavenumber from 2000 cm−1 to 2300 cm−1 or a Fourier transform-infrared spectroscopy spectroscopy peak with wavenumber from 2000 cm−1 to 2300 cm−1, and also to methods of DNA sequencing and SNP detection.

    摘要翻译: 本发明提供了具有以下结构的核苷三磷酸类似物:其中B为碱基,为腺嘌呤,鸟嘌呤,胞嘧啶,尿嘧啶或胸腺嘧啶,其中R“为OH或H,并且其中R'为叠氮基甲基,烃基或取代 烃基,并且具有波长为2000cm -1至2300cm -1的拉曼光谱峰或波数为2000cm -1至2300cm -1的傅立叶变换红外光谱光谱峰,以及DNA测序方法 和SNP检测。

    Methods of manufacturing light to current converter devices
    8.
    发明授权
    Methods of manufacturing light to current converter devices 有权
    制造光电转换器装置的方法

    公开(公告)号:US08916410B2

    公开(公告)日:2014-12-23

    申请号:US13193458

    申请日:2011-07-28

    摘要: Processes for making light to current converter devices are provided. The processes can be used to make light to current converter devices having P-N junctions located on only the top surface of the cell, located on the top surface and symmetrically or asymmetrically along a portion of the inner surface of the via holes, located on the top surface and full inner surface of the via holes, or located on the top surface, full inner surface of the via holes, and a portion of the bottom surface of the cell. The processes may isolate the desired P-N junction by etching the emitter, forming a via hole after forming the emitter, using a barrier layer to protect portions of the emitter from etching, or using a barrier layer to prevent the emitter from being formed on portions of the substrate.

    摘要翻译: 提供了制造光电流转换器装置的过程。 这些工艺可以用于对位于顶表面上的仅位于电池顶表面上的PN结的电流转换器装置进行光照,并且沿着位于顶部的通孔的内表面的一部分对称地或不对称地 通孔的表面和整个内表面,或位于顶表面,通孔的完整内表面和电池底表面的一部分。 这些过程可以通过蚀刻发射体来隔离所需的PN结,在形成发射体之后使用阻挡层来保护发射体的部分免受蚀刻,或者使用阻挡层来防止发射体在 底物。

    FULL-AUTOMATIC DETECTING SYSTEM AND METHOD FOR TRANSFORMER
    10.
    发明申请
    FULL-AUTOMATIC DETECTING SYSTEM AND METHOD FOR TRANSFORMER 有权
    全自动检测系统及变压器的方法

    公开(公告)号:US20130297245A1

    公开(公告)日:2013-11-07

    申请号:US13980556

    申请日:2011-11-25

    IPC分类号: G01N35/00

    摘要: A full-automatic detecting system and method for a transformer. The system comprises a material delivery line; a feed device; a detection delivery line; insulating test devices; error detecting devices; a laser marking device; a discharge device; and a detection management system. The insulating test devices, the error detecting devices, and the laser marking device are sequentially arranged beside the detection delivery line. The material delivery line, the detection delivery line, and the devices work in coordination, to realize full-automatic detection of the transformer, avoid manual detection errors and improve the sorting accuracy.

    摘要翻译: 一种变压器全自动检测系统及方法。 该系统包括物料输送管线; 饲料装置; 检测传输线; 绝缘试验装置; 检错装置; 激光打标装置; 放电装置; 和检测管理系统。 隔离测试装置,误差检测装置和激光打标装置依次布置在检测输送线的旁边。 物料输送线,检测输送线和设备协调工作,实现变压器全自动检测,避免手动检测错误,提高分拣精度。