公开(公告)号：US20150117652A1

公开(公告)日：2015-04-30

申请号：US14404500

申请日：2012-05-31

申请人： Jun Sato ,  Ryuji Funayama ,  Tomoya Takatani ,  Toshiki Kindo ,  Hideo Fukamachi

发明人： Jun Sato ,  Ryuji Funayama ,  Tomoya Takatani ,  Toshiki Kindo ,  Hideo Fukamachi

IPC分类号： H04R29/00 ,  G10K11/00

CPC分类号： H04R29/00 ,  G10K11/002 ,  G10L21/0216 ,  G10L21/0232 ,  G10L25/78 ,  G10L2021/02166

摘要： Disclosed is a noise model generation device that generates a noise model suitable for each environment by determining whether an sound source of a detection target is included in sound information collected by an sound collector with high accuracy. The noise model generation device generates a noise model relating to noise information other than the sound source of the detection target included in the sound information collected by the sound collector, which acquires a power spectrum from the sound information; determines whether the sound source of the detection target is included in the sound information collected by evaluating a probability density distribution (histogram) of the power spectrum; and generates a noise model from the collected sound information when it is determined that the sound source of the detection target is not included in the collected sound information.

摘要翻译： 公开了一种噪声模型生成装置，其通过确定检测对象的声源是否被包含在由收集器以高精度收集的声音信息中，从而生成适合于每个环境的噪声模型。 噪声模型生成装置生成与声音收集器收集的声音信息中包含的检测对象的声源以外的噪声信息相关的噪声模型，该声音信息从声音信息获取功率谱; 通过评估功率谱的概率密度分布（直方图）来确定检测对象的声源是否包含在收集的声音信息中; 并且当确定检测目标的声源不包括在所收集的声音信息中时，从所收集的声音信息中产生噪声模型。

公开(公告)号：US09299334B2

公开(公告)日：2016-03-29

申请号：US14346060

申请日：2012-09-19

申请人： Jun Sato ,  Ryuji Funayama ,  Tomoya Takatani ,  Osami Yamamoto ,  Kensaku Asahi ,  Hideki Banno ,  Keiichi Yamada ,  Akira Ogawa ,  Yuichi Tanaka ,  Hiroyuki Hoshino

发明人： Jun Sato ,  Ryuji Funayama ,  Tomoya Takatani ,  Osami Yamamoto ,  Kensaku Asahi ,  Hideki Banno ,  Keiichi Yamada ,  Akira Ogawa ,  Yuichi Tanaka ,  Hiroyuki Hoshino

IPC分类号： G01S1/14 ,  G10K11/00 ,  G01S3/808 ,  G01S3/86 ,  H04R29/00

CPC分类号： G10K11/002 ,  G01S3/8083 ,  G01S3/86 ,  H04R29/00

摘要： A sound source detecting system that detects a predetermined sound source on the basis of sounds collected by sound collectors includes: a noise extracting unit that extracts a noise from signals of the collected sounds; and a noise suppressing unit that suppresses a signal component of the noise extracted by the noise extracting unit from the signals of the collected sounds. The sound source detecting unit detects a location of the predetermined sound source, such as an approaching vehicle, using information about the sounds having the signals of which the noise is suppressed.

摘要翻译： 基于由收集器收集的声音来检测预定声源的声源检测系统包括：噪声提取单元，从收集的声音的信号中提取噪声; 以及噪声抑制单元，其从收集的声音的信号中抑制由噪声提取单元提取的噪声的信号分量。 声源检测单元使用关于具有噪声被抑制的信号的声音的信息来检测诸如接近的车辆的预定声源的位置。

公开(公告)号：US20180166077A1

公开(公告)日：2018-06-14

申请号：US15834129

申请日：2017-12-07

申请人： Toru Yamaguchi ,  Eri Shimokawara ,  Kazuki Hirata ,  Tomoya Takatani

发明人： Toru Yamaguchi ,  Eri Shimokawara ,  Kazuki Hirata ,  Tomoya Takatani

IPC分类号： G10L15/22 ,  G10L13/00 ,  G10L15/06 ,  G10L15/14 ,  G06F17/30

CPC分类号： G10L15/22 ,  G06F16/634 ,  G06F17/2785 ,  G10L13/00 ,  G10L15/063 ,  G10L15/14 ,  G10L2015/223 ,  H04L51/02

摘要： A language storage method according to one aspect of the present invention includes vectorization steps for vectorizing language accumulated in a dialog log database using a word vector created in advance for each word; determination step for determining whether to employ the language that has been vectorized as response data; and a storage step for accumulating the language that has been determined to be employed as the response data in a response database, in which a dimension reduction step for reducing the number of dimensions of the vector of the language vectorized in the vectorization step is included between the vectorization step and the determination step, and in the determination step, it is determined which one of an employed group and an unemployed group the vector of the language whose number of dimensions has been reduced in the dimension reduction step belongs to.

公开(公告)号：US08755537B2

公开(公告)日：2014-06-17

申请号：US13177219

申请日：2011-07-06

申请人： Tomoya Takatani

发明人： Tomoya Takatani

IPC分类号： H04R3/00 ,  H04R29/00 ,  H04R25/00

CPC分类号： H04R1/406 ,  G01H1/00 ,  H04R3/005 ,  H04R2499/13

摘要： To provide a microphone unit capable of acquiring a target sound with high accuracy. A microphone unit in accordance with an exemplary embodiment of the present invention includes a plurality of microphones, a microphone substrate on which the plurality of microphones are mounted, and a vibration observation device disposed at roughly a center of gravity of a shape that is formed by connecting centers of certain adjacent microphones among the plurality of microphones.

摘要翻译： 提供能够高精度地获取目标声音的麦克风单元。 根据本发明的示例性实施例的麦克风单元包括多个麦克风，安装有多个麦克风的麦克风基板，以及大致重心形成的振动观察装置，其形状形成为： 在多个麦克风之间连接某些相邻麦克风的中心。

公开(公告)号：US08452592B2

公开(公告)日：2013-05-28

申请号：US12921974

申请日：2008-09-02

申请人： Tomoya Takatani ,  Jani Even

发明人： Tomoya Takatani ,  Jani Even

IPC分类号： G10L21/02

CPC分类号： H04S1/007 ,  G10L21/0272

摘要： Provided are a signal separating apparatus and a signal separating method capable of solving the permutation problem and separating user speech to be extracted. The signal separating apparatus separates a specific speech signal and a noise signal from a received sound signal. First, a joint probability density distribution estimation unit of a permutation solving unit calculates joint probability density distributions of the respective separated signals. Then, a classifying determination unit of the permutation solving unit determines classifying based on shapes of the calculated joint probability density distributions.

摘要翻译： 提供了能够解决置换问题并分离要提取的用户语音的信号分离装置和信号分离方法。 信号分离装置从接收到的声音信号中分离特定语音信号和噪声信号。 首先，置换求解单元的联合概率密度分布估计单元计算各个分离信号的联合概率密度分布。 然后，置换求解单元的分类确定单元基于所计算的联合概率密度分布的形状来确定分类。

公开(公告)号：US20110152709A1

公开(公告)日：2011-06-23

申请号：US13060082

申请日：2009-09-18

申请人： Hitoshi Yamada ,  Tomoya Takatani ,  Kazutoshi Ebe ,  Andrzej Cichocki ,  Kyuwan Choi

发明人： Hitoshi Yamada ,  Tomoya Takatani ,  Kazutoshi Ebe ,  Andrzej Cichocki ,  Kyuwan Choi

IPC分类号： A61B5/0476

CPC分类号： G06F3/015 ,  A61G5/04 ,  A61G2203/18 ,  B60Y2200/84 ,  G06K9/00496

摘要： A mobile body control device includes: a brain activity detecting unit that detects brain activity information of a user; a brain signal separating unit that separates an artifact component from the brain activity information detected by the brain activity detecting unit; a control signal generating unit that slides a sampling period for extracting brain data, at predetermined intervals in an overlapped manner in the brain activity information from which the artifact component is separated, successively calculates feature values for the brain data within each of the sampling periods obtained by sliding, and generates a control signal based on the feature values calculated; and a drive control unit that drives and controls a mobile body with a user riding thereon, based on the control signal generated.

摘要翻译： 移动体控制装置包括：大脑活动检测单元，其检测用户的大脑活动信息; 脑信号分离单元，其将神经元成分与由大脑活动检测单元检测出的大脑活动信息分离; 一个控制信号产生单元，用于以预先确定的时间间隔以预定的间隔滑动用于提取大脑数据的采样周期，所述大脑活动信息从其中分离出的神经元分量连续地计算所获取的每个采样周期内的大脑数据的特征值 并基于所计算的特征值产生控制信号; 以及驱动控制单元，其基于所生成的控制信号，驱动并控制其上搭载有用户的移动体。

公开(公告)号：US20110029309A1

公开(公告)日：2011-02-03

申请号：US12921974

申请日：2008-09-02

申请人： Tomoya Takatani ,  Jani Even

发明人： Tomoya Takatani ,  Jani Even

IPC分类号： G10L15/20

CPC分类号： H04S1/007 ,  G10L21/0272

摘要： Provided are a signal separating apparatus and a signal separating method capable of solving the permutation problem and separating user speech to be extracted. The signal separating apparatus separates a specific speech signal and a noise signal from a received sound signal. First, a joint probability density distribution estimation unit of a permutation solving unit calculates joint probability density distributions of the respective separated signals. Then, a classifying determination unit of the permutation solving unit determines classifying based on shapes of the calculated joint probability density distributions.

摘要翻译： 提供了能够解决置换问题并分离要提取的用户语音的信号分离装置和信号分离方法。 信号分离装置从接收到的声音信号中分离特定语音信号和噪声信号。 首先，置换求解单元的联合概率密度分布估计单元计算各个分离信号的联合概率密度分布。 然后，排列求解单元的分类确定单元基于计算的联合概率密度分布的形状来确定分类。