利索-全球专利检索

  1. 登录
  2. 注册

搜索结果

11 条记录 (耗时 0.021 秒)

    Method and apparatus for reducing noise corruption from an alternative sensor signal during multi-sensory speech enhancement
    2.
    发明授权
    Method and apparatus for reducing noise corruption from an alternative sensor signal during multi-sensory speech enhancement 有权
    用于在多感觉语音增强期间从替代传感器信号减少噪声损坏的方法和装置

    公开(公告)号：US07590529B2

    公开(公告)日：2009-09-15

    申请号：US11050936

    申请日：2005-02-04

    申请人： Zhengyou Zhang ,  Amarnag Subramanya ,  James G. Droppo ,  Zicheng Liu

    发明人： Zhengyou Zhang ,  Amarnag Subramanya ,  James G. Droppo ,  Zicheng Liu

    IPC分类号： G10L21/02 ,  G10L15/20

    CPC分类号： G10L21/0208 ,  G10L2021/02165

    摘要： A method and apparatus classify a portion of an alternative sensor signal as either containing noise or not containing noise. The portions of the alternative sensor signal that are classified as containing noise are not used to estimate a portion of a clean speech signal and the channel response associated with the alternative sensor. The portions of the alternative sensor signal that are classified as not containing noise are used to estimate a portion of a clean speech signal and the channel response associated with the alternative sensor.

    摘要翻译： 方法和装置将替代传感器信号的一部分分类为包含噪声或不包含噪声。 分类为包含噪声的替代传感器信号的部分不用于估计清洁语音信号的一部分和与替代传感器相关联的信道响应。 被分类为不包含噪声的替代传感器信号的部分被用于估计清洁语音信号的一部分和与替代传感器相关联的信道响应。

    Multi-sensory speech enhancement using synthesized sensor signal
    6.
    发明授权
    Multi-sensory speech enhancement using synthesized sensor signal 有权
    使用合成传感器信号的多感觉语音增强

    公开(公告)号：US07406303B2

    公开(公告)日：2008-07-29

    申请号：US11228710

    申请日：2005-09-16

    申请人： Li Deng ,  Zhengyou Zhang ,  Zicheng Liu ,  Amarnag Subramanya

    发明人： Li Deng ,  Zhengyou Zhang ,  Zicheng Liu ,  Amarnag Subramanya

    IPC分类号： H04B1/06 ,  H04B7/00

    CPC分类号： G10L21/0208 ,  G10L2021/02165

    摘要： A synthesized alternative sensor signal is produced from an alternative sensor signal. The synthesized alternative sensor signal is computed using vocal tract resonances estimated based on the alternative sensor signal, and using a waveform synthesis technique that converts the estimated vocal tract resonance sequence into a spectral magnitude sequence. The synthesized alternative sensor signal and the alternative sensor signal are used to estimate a clean speech value.

    摘要翻译： 从替代传感器信号产生合成的替代传感器信号。 使用基于替代传感器信号估计的声道共振来计算合成的替代传感器信号，并且使用将估计的声道共振序列转换为频谱幅度序列的波形合成技术。 合成的替代传感器信号和替代传感器信号用于估计干净的语音值。

    Speech modeling and enhancement based on magnitude-normalized spectra
    7.
    发明申请
    Speech modeling and enhancement based on magnitude-normalized spectra 有权
    基于幅度归一化光谱的语音建模和增强

    公开(公告)号：US20070150263A1

    公开(公告)日：2007-06-28

    申请号：US11317269

    申请日：2005-12-23

    申请人： Zhengyou Zhang ,  Alejandro Acero ,  Amarnag Subramanya ,  Zicheng Liu

    发明人： Zhengyou Zhang ,  Alejandro Acero ,  Amarnag Subramanya ,  Zicheng Liu

    IPC分类号： G10L21/00

    CPC分类号： G10L21/0208 ,  G10L15/20

    摘要： A frame of a speech signal is converted into the spectral domain to identify a plurality of frequency components and an energy value for the frame is determined. The plurality of frequency components is divided by the energy value for the frame to form energy-normalized frequency components. A model is then constructed from the energy-normalized frequency components and can be used for speech recognition and speech enhancement.

    摘要翻译： 语音信号的帧被转换成频谱域以识别多个频率分量，并确定该帧的能量值。 将多个频率分量除以帧的能量值以形成能量归一化频率分量。 然后从能量归一化的频率分量构建模型，并可用于语音识别和语音增强。

    Multi-sensory speech enhancement using synthesized sensor signal
    8.
    发明申请
    Multi-sensory speech enhancement using synthesized sensor signal 有权
    使用合成传感器信号的多感觉语音增强

    公开(公告)号：US20070010291A1

    公开(公告)日：2007-01-11

    申请号：US11228710

    申请日：2005-09-16

    申请人： Li Deng ,  Zhengyou Zhang ,  Zicheng Liu ,  Amarnag Subramanya

    发明人： Li Deng ,  Zhengyou Zhang ,  Zicheng Liu ,  Amarnag Subramanya

    IPC分类号： H04B1/00 ,  H04B1/38 ,  H04B1/10

    CPC分类号： G10L21/0208 ,  G10L2021/02165

    摘要： A synthesized alternative sensor signal is produced from an alternative sensor signal. The synthesized alternative sensor signal is computed using vocal tract resonances estimated based on the alternative sensor signal, and using a waveform synthesis technique that converts the estimated vocal tract resonance sequence into a spectral magnitude sequence. The synthesized alternative sensor signal and the alternative sensor signal are used to estimate a clean speech value.

    摘要翻译： 从替代传感器信号产生合成的替代传感器信号。 使用基于替代传感器信号估计的声道共振来计算合成的替代传感器信号，并且使用将估计的声道共振序列转换为频谱幅度序列的波形合成技术。 合成的替代传感器信号和替代传感器信号用于估计干净的语音值。

    Graph-based semi-supervised learning of structured tagging models
    10.
    发明授权
    Graph-based semi-supervised learning of structured tagging models 有权
    结构化标注模型的基于图形的半监督学习

    公开(公告)号：US08560477B1

    公开(公告)日：2013-10-15

    申请号：US12901356

    申请日：2010-10-08

    申请人： Slav Petrov ,  Amarnag Subramanya ,  Fernando Pereira ,  Dipanjan Das

    发明人： Slav Petrov ,  Amarnag Subramanya ,  Fernando Pereira ,  Dipanjan Das

    IPC分类号： G06F17/00 ,  G06N5/02

    CPC分类号： G06N99/005 ,  G06K9/6224 ,  G06K9/72 ,  G06K2209/01

    摘要： Methods, systems, and apparatus, including computer programs encoded on a computer storage medium, for graph-based semi-supervised learning of structured tagging models. In one aspect, a method includes creating a graph having a plurality of unique vertices in which vertices in a first set of vertices represent n-grams that are each associated with a respective part-of-speech and that were derived from labeled source domain text, and in which vertices in a different second set of vertices represent n-grams that are not associated with a part-of-speech and that were derived from unlabeled target domain text. A respective measure of similarity is calculated between the vertices in each of the pairs based at least partially on a distance between the respective features of the pair used to weight a graph edge between the pair.

    摘要翻译： 方法，系统和装置，包括在计算机存储介质上编码的计算机程序，用于基于图形的结构化标记模型的半监督学习。 在一个方面，一种方法包括创建具有多个唯一顶点的图形，其中第一组顶点中的顶点表示n个克，其各自与相应的词性相关联并且从标记的源域文本导出 ，并且不同第二组顶点中的顶点表示与词性不相关并且是从未标记的目标域文本导出的n-gram。 至少部分地基于用于加权该对之间的图形边缘的对的各个特征之间的距离，在每对中的顶点之间计算相似度的相应度量。