利索-全球专利检索

  1. 登录
  2. 注册

搜索结果

22 条记录 (耗时 0.036 秒)

    Channel normalization apparatus and method for robust speech recognition
    7.
    发明授权
    Channel normalization apparatus and method for robust speech recognition 失效
    用于鲁棒语音识别的信道归一化装置和方法

    公开(公告)号：US07702505B2

    公开(公告)日：2010-04-20

    申请号：US11300223

    申请日：2005-12-14

    申请人： Ho-Young Jung

    发明人： Ho-Young Jung

    IPC分类号： G10L19/14

    CPC分类号： G10L15/20

    摘要： A channel normalization apparatus includes: a characteristic extraction unit extracting MFCC characteristics and outputting rows of frames according to time; a characteristic parameter average calculation unit calculating an average value of the rows of the outputted MFCC characteristics; a channel variation estimation unit configuring a codebook based on a database of speech signals with attenuated channel variations and estimating a channel variation for each frame by calculating a distance between a MFCC parameter for each frame and an individual median value of the codebook when a MFCC of a channel distorted speech signal is inputted; and a smoothing operation based channel normalization unit smoothing another average value of the channel variation from the characteristic parameter average calculation unit and the channel variation from the channel variation estimation unit, subtracting the other average value from the MFCC of each frame and outputting rows of channel normalized MFCC characteristics.

    摘要翻译： 信道归一化装置包括：特征提取单元，提取MFCC特性并根据时间输出帧帧; 特征参数平均计算单元，计算输出的MFCC特性的行的平均值; 信道变化估计单元，基于具有衰减信道变化的语音信号的数据库构建码本，并且通过计算每帧的MFCC参数与码本的MFCC参数之间的距离，并且当MFCC的MFCC为 输入信道失真语音信号; 以及基于平滑运算的信道归一化单元，从所述特征参数平均计算单元对所述信道变化的另一平均值和来自所述信道变化估计单元的信道变化进行平滑，从每帧的MFCC减去另一平均值，并输出信道行 归一化MFCC特性。

    SHUTTER DEVICE FOR CAMERA
    9.
    发明申请
    SHUTTER DEVICE FOR CAMERA 失效
    摄像机快门设备

    公开(公告)号：US20110038626A1

    公开(公告)日：2011-02-17

    申请号：US12988212

    申请日：2008-12-16

    申请人： Dae-Soon Lim ,  Kyung-Ho Kim ,  Jin-Chul Ahn ,  Tae-Jin Kim ,  Ho-Young Jung

    发明人： Dae-Soon Lim ,  Kyung-Ho Kim ,  Jin-Chul Ahn ,  Tae-Jin Kim ,  Ho-Young Jung

    IPC分类号： G03B9/08 ,  H04N5/235

    CPC分类号： G03B9/08

    摘要： The present invention relates to a shutter device for camera used for small-sized digital equipment and includes a base forming a light transmission hole to expose a film or charge-coupled device (CCD) to laser beam; an electromagnet mounted on one side of the base; a driving arm which is mounted on one side of the electromagnet on the base so as to pivot on the base; a first shutter and a second shutter to open and close the light transmission hole, which pivots on the part of the base in association with the pivoting of the driving arm; and a magnetic which is installed on the driving arm and has the north magnetic pole and the south magnetic pole at opposite ends of the magnetic or on surfaces opposite to the electromagnet. The driving arm drives the first shutter and the second shutter by pivoting in a clockwise or counter-clockwise direction by the attractive force and repulsive force generated between the electromagnet and the magnetic according to the direction of the electric current applied to the electromagnet, the electromagnet includes a bobbin and a coil wound around the bobbin, and each side of the bobbin having opposite poles of magnetic field is disposed adjacent to the north and south pole.

    摘要翻译： 本发明涉及一种用于小型数字设备的照相机快门装置，包括形成透光孔的底座，以将胶片或电荷耦合装置（CCD）暴露于激光束; 安装在基座的一侧的电磁体; 驱动臂，其安装在基座上的电磁体的一侧，以便在基座上枢转; 第一快门和第二快门，用于打开和关闭光传输孔，该光传输孔与驱动臂的枢转相关联地在基座的一部分上枢转; 以及安装在驱动臂上并具有北磁极的南极磁极和与磁铁相对的表面的相对端的磁铁。 驱动臂根据施加到电磁铁的电流的方向，通过在电磁体和磁体之间产生的吸引力和排斥力沿顺时针或逆时针方向枢转来驱动第一快门和第二快门，电磁体 包括卷绕在绕线架上的绕线管和线圈，并且具有与磁极相反的磁极的线轴的每一侧设置为与北极和南极相邻。

    SPEECH RECOGNITION SYSTEM AND METHOD
    10.
    发明申请
    SPEECH RECOGNITION SYSTEM AND METHOD 有权
    语音识别系统和方法

    公开(公告)号：US20100161326A1

    公开(公告)日：2010-06-24

    申请号：US12506705

    申请日：2009-07-21

    申请人： Sung Joo Lee ,  Ho-Young Jung ,  Jeon Gue Park ,  Hoon Chung ,  Yunkeun Lee ,  Byung Ok Kang ,  Hyung-Bae Jeon ,  Jong Jin Kim ,  Ki-young Park ,  Euisok Chung ,  Ji Hyun Wang ,  Jeom Ja Kang

    发明人： Sung Joo Lee ,  Ho-Young Jung ,  Jeon Gue Park ,  Hoon Chung ,  Yunkeun Lee ,  Byung Ok Kang ,  Hyung-Bae Jeon ,  Jong Jin Kim ,  Ki-young Park ,  Euisok Chung ,  Ji Hyun Wang ,  Jeom Ja Kang

    IPC分类号： G10L15/20

    CPC分类号： G10L21/0208 ,  G10L15/20 ,  G10L25/48

    摘要： A speech recognition system includes: a speed level classifier for measuring a moving speed of a moving object by using a noise signal at an initial time of speech recognition to determine a speed level of the moving object; a first speech enhancement unit for enhancing sound quality of an input speech signal of the speech recognition by using a Wiener filter, if the speed level of the moving object is equal to or lower than a specific level; and a second speech enhancement unit enhancing the sound quality of the input speech signal by using a Gaussian mixture model, if the speed level of the moving object is higher than the specific level. The system further includes an end point detection unit for detecting start and end points, an elimination unit for eliminating sudden noise components based on a sudden noise Gaussian mixture model.

    摘要翻译： 语音识别系统包括：速度级分类器，用于通过在语音识别的初始时间使用噪声信号来测量移动物体的移动速度，以确定移动物体的速度水平; 第一语音增强单元，如果移动对象的速度水平等于或低于特定水平，则通过使用维纳滤波器来增强语音识别的输入语音信号的声音质量; 以及第二语音增强单元，如果移动对象的速度水平高于特定水平，则通过使用高斯混合模型来增强输入语音信号的声音质量。 该系统还包括用于检测起点和终点的终点检测单元，用于基于突发噪声高斯混合模型消除突发噪声分量的消除单元。