-
公开(公告)号：KR101442700B1
公开(公告)日：2014-09-23
申请号：KR1020130105730
申请日：2013-09-03
Applicant: 서강대학교산학협력단
CPC classification number: H04R3/04 , H04R1/1083 , H04R2410/05
Abstract: A noise cancelation apparatus for headphones using an independent component analysis adaptive filter includes; an external microphone installed at an outside of the headphones and to receive an external noise; a feedforward adaptive filter using independent component analysis which receives a noise signal output from the external microphone to perform the convolution for an adaptive filter coefficient using the noise signal as a reference signal for generating an attenuation signal; a subtracter which subtracts the attenuation signal from a target signal provided from the outside; and an error microphone installed inside the headphones. An output signal is outputted through a speaker of the headphones. The adaptive filter coefficient is learned by receiving an error signal from the error microphone to minimize a component which is not correlated to each other between the target signal and the noise signal.
Abstract translation: 使用独立分量分析自适应滤波器的耳机噪声消除装置包括： 安装在耳机外部并接收外部噪声的外部麦克风; 使用独立分量分析的前馈自适应滤波器，其接收从外部麦克风输出的噪声信号，以使用噪声信号作为用于产生衰减信号的参考信号来执行自适应滤波器系数的卷积; 从外部提供的目标信号中减去衰减信号的减法器; 并安装在耳机内的错误麦克风。 通过耳机的扬声器输出输出信号。 通过从误差麦克风接收误差信号来学习自适应滤波器系数，以使在目标信号和噪声信号之间彼此不相关的分量最小化。
-
公开(公告)号：KR1020130068869A
公开(公告)日：2013-06-26
申请号：KR1020110136285
申请日：2011-12-16
Applicant: 서강대학교산학협력단
CPC classification number: H04R3/02 , G10L15/20 , G10L21/0208 , G10L21/0272 , H04R3/005 , H04R2410/03
Abstract: PURPOSE: A method for removing an interested sound source and a method for recognizing a voice are provided to estimate a mixed noise signal effectively, by removing the interested sound source from a mixed signal provided through two microphones installed in an acoustic signal mixed environment. CONSTITUTION: A voice recognition device initializes a vector(12). The device learns the vector to remove an interested sound source from a mixed signal(10). The device is initialized, and generates a mixed noise signal by removing an interested sound source signal from an input mixed signal(14). When the mixed noise signal is generated, the device generates a mask, by comparing the mixed noise signal with the input mixed signal in a time-frequency domain(16). [Reference numerals] (10) Learn a separation vector(w(K)) for separation of a target sound source; (12) Initialize the separation vector(w(K)); (14) Remove a target sound source; (16) Generate a mask using a mixed noise sound source and an input sound source
Abstract translation: 目的：提供一种消除感兴趣的声源的方法和用于识别语音的方法，通过从通过安装在声信号混合环境中的两个麦克风提供的混合信号中去除感兴趣的声源来有效地估计混合噪声信号。 构成：语音识别装置初始化矢量（12）。 该装置学习矢量以从混合信号中去除感兴趣的声源（10）。 该装置被初始化，并且通过从输入混合信号（14）中去除感兴趣的声源信号来产生混合噪声信号。 当产生混合噪声信号时，器件通过将混合噪声信号与时频域（16）中的输入混合信号进行比较来产生掩模。 （附图标记）（10）学习用于分离目标声源的分离矢量（w（K））; （12）初始化分离向量（w（K））; （14）移除目标声源; （16）使用混合噪声声源和输入声源产生掩模
-
公开(公告)号：KR101260684B1
公开(公告)日：2013-05-10
申请号：KR1020110051512
申请日：2011-05-30
Applicant: 서강대학교산학협력단
IPC: G10L19/00
Abstract: 본발명의블라인드소스분리장치는, 혼합신호들을 FD의혼합신호들로변환하는 ST(Short-Time) 푸리에트랜스포밍하는 ST(Short-Time) 푸리에트랜스포머; 상기 FD의혼합신호들을소스신호들로분리하는 FF(Feed Forward) 언믹싱필터네트워크; 상기 FD의분리된소스신호들을 TD로되돌리는역 ST(Short-Time) 푸리에트랜스포머; 를포함하는것을특징으로한다.
-
公开(公告)号：KR1020130006857A
公开(公告)日：2013-01-18
申请号：KR1020110061697
申请日：2011-06-24
Applicant: 서강대학교산학협력단
Abstract: PURPOSE: Time delay of signal and a mute sound source dividing method in repercussion environment based on attenuation estimation are provided to consider repercussion environment, estimate per-frequency attenuation and time delay value, and divide mute sound source signal, thereby improve dividing function of the mute sound source signal. CONSTITUTION: Mute sound source divider performs initialization for per-frequency attenuation and time delay value for inputted mixed signal(200). The mute sound source divider learns the per-frequency attenuation and time delay value(202). The mute sound source divider divides per-frequency signal by usage of per-frequency binary mask(206). The mute sound source divider matches order by acquisition of coefficient of correlation of the divided signal(208). The mute sound source divider optimizes the order(210). [Reference numerals] (200) Initializing a per-frequency attenuation and time delay value; (202) Learning a per-frequency attenuation and time delay value; (204) Generating a per-frequency binary mask; (206) Dividing a per-frequency sound source signal by using a per-frequency binary mask; (208) Arranging orders by obtaining correlation coefficient for each per-frequency dividing signal; (210) Order optimization; (AA) Start; (BB) End
Abstract translation: 目的：提供基于衰减估计的反响环境中信号和静音声源分割方法的延时，考虑影响环境，估计每频衰减和时间延迟值，并分割静音声源信号，从而提高分频功能 静音声源信号。 构成：静音声源分频器对输入的混合信号（200）的每频率衰减和时间延迟值进行初始化。 静音声源分频器了解每频衰减和时间延迟值（202）。 静音声源分频器通过使用每频率二进制掩码（206）来划分每频率信号。 静音声源分频器通过获取分频信号的相关系数（208）来匹配顺序。 静音声源分频器优化顺序（210）。 （附图标记）（200）初始化每频率衰减和时间延迟值; （202）学习频率衰减和时间延迟值; （204）生成全频二进制掩码; （206）通过使用全频二进制掩码来划分每频率声源信号; （208）通过获得每个分频信号的相关系数来排列顺序; （210）订单优化; （AA）开始; （BB）结束
-
-
Patent Agency Ranking
-
-
公开(公告)号：KR101899071B1
公开(公告)日：2018-09-14
申请号：KR1020160184132
申请日：2016-12-30
Applicant: 서강대학교산학협력단
Abstract: 본발명의실시예에따른영상에서전경영역을국소화하는방법은, 영상을입력받는영상입력단계와, 영상에서회색또는컬러가우스혼합모델에기반을둔 배경혼합모델을사용하여다수의전경후보픽셀을추출하는단계와, 다수의전경후보픽셀들에공간적가우스혼합모델피팅을수행하여공간적가우스혼합모델을생성하는단계와, 공간적가우스혼합모델의성분을각 성분의비중의내림차순으로정렬하는단계및 공간적가우스혼합모델의성분을정렬된순서에따라차례로, 평균이가까운두 성분을병합하는과정을마지막성분에까지반복수행하는단계를포함하여구성된다.
-
公开(公告)号：KR101840015B1
公开(公告)日：2018-04-26
申请号：KR1020160175302
申请日：2016-12-21
Applicant: 서강대학교산학협력단
CPC classification number: G10H1/366 , G10H1/0008 , G10H2210/056 , G10H2210/305 , G10H2250/135
Abstract: 본발명에따르는스테레오음악신호를위한반주신호추출장치는, 스테레오음악신호를입력받아소프트마스크를상기시간-주파수영역의음악신호에적용하여노래음성신호와백그라운드반주신호를분리하여출력하는패닝프로세싱부; 상기패닝프로세싱부가출력하는노래음성신호를저해상도스펙트럼으로변환하여크기스펙트럼을기준으로메디안필터링하여하모닉및 타악신호를추출하는하모닉및 타악신호분리부; 및상기백그라운드반주신호와상기하모닉및 타악신호를입력받아가산하여강화된반주신호를생성하여출력하는가산부;로구성됨을특징으로한다.
-
公开(公告)号：KR101721424B1
公开(公告)日：2017-03-31
申请号：KR1020150191321
申请日：2015-12-31
Applicant: 서강대학교산학협력단
Abstract: 본발명은반향에강인한다음원위치탐지방법에관한것이다. 상기다음원위치탐지방법은, (a) 다수개의마이크를통해입력된관찰신호를사전설정된길이단위로블록화하는단계; (b) 관찰신호의임의의블록에대하여주파수영역의신호로변환시키는단계; (c) 주파수영역의신호로변환된관찰신호의해당블록에대하여, 주파수영역에서의우도함수를설정하고상기우도함수를최대화하는방향으로채널필터를학습하는단계; (d) 관찰신호의해당블록에대하여상기학습된채널필터를이용하여음원신호를추정하고, 추정된음원신호에대하여서로소조건을적용한후, 채널필터를업데이트시키는단계; (e) 관찰신호의해당블록에대하여상기업데이트된채널필터를시간영역의신호로변환시킨후 성김조건을적용하는단계; (f) 관찰신호의해당블록에대한학습된채널필터를정규화시키는단계; (g) 상기관찰신호의각 블록들에대하여상기 (b) 단계내지 (f) 단계를반복수행하여채널필터를학습하는단계;를구비하여, 상기학습된채널필터를이용하여음원의위치를탐지해낼수 있다.
Abstract translation: 本发明涉及一种回声抵抗下一个原位检测方法。 下一个原位检测方法包括以下步骤：（a）将通过多个麦克风输入的观测信号阻挡预定长度单位; （b）将观测信号的任意块变换成频域信号; （C）设置一个似然函数eseoui，所观察到的信号的相应块中的频域转换为频域信号，并以最大化似然函数的方向上的学习过滤器通道; （d）使用所学习的信道滤波器，更新估计声源信号，相对于所估计的声源信号，对观测信号的相应块中的信道滤波器施加预定条件彼此之后的步骤; （e）将更新的信道滤波器变换成用于观测信号的相应块的时域信号，并应用渲染条件; （f）针对观测信号的该块对学习的信道滤波器进行归一化; 检测所述声源的位置并设置有，使用学习的信道滤波器;（G）通过重复执行步骤（b）至（f）步骤的观测信号的每个块学习信道滤波器的步骤 我可以做到。
-
-
-
-
-
-
-
-
-
Search Results
72
records
(took 0.024 seconds)
