专利快速检索-快速检索全球专利，免费商用专利数据库-IPRDB

1. 发明专利

JP2017083566A 雑音抑圧装置、雑音抑圧方法、およびプログラム审中-公开
公开(公告)号：JP2017083566A
公开(公告)日：2017-05-18
申请号：JP2015209729
申请日：2015-10-26
申请人：日本電信電話株式会社 , Nippon Telegr & Teleph Corp
发明人： KAWASE TOMOKO , NIWA KENTA , FUJIMOTO MASAKIYO , KAMADO NORIYOSHI , NAKATANI TOMOHIRO , ARAKI AKIKO , KOBAYASHI KAZUNORI
IPC分类号： G10L21/0232 , G10L21/0208 , G10L21/0264
摘要：【課題】雑音抑圧性能を高める。【解決手段】モデルパラメータ記憶部10はクリーン音声信号および雑音信号の少なくともいずれかと無音信号を混合正規分布で表現した信号成分モデルのパラメータを記憶する。ビームフォーマー部12はマイクロホンアレー11の出力信号を利用して角度領域信号を生成する。周波数領域変換部13は角度領域信号を周波数領域信号に変換する。特定方向選択部14は特定方向周波数領域信号を選択する。特定方向信号モデル化部17は混合正規分布で表現される特定方向信号モデルを生成し、特定方向周波数領域信号が観測されたときの特定方向信号モデルの事後確率を計算する。利得係数算出部18は利得係数を特定方向信号モデルの事後確率で重み付き加算する。乗算部19は利得係数を特定方向周波数領域信号に乗算する。【選択図】図1

2. 发明专利

JP2013097176A Sound source separation device, sound source separation method, and program 有权
标题翻译：声源分离装置，声源分离方法和程序
公开(公告)号：JP2013097176A
公开(公告)日：2013-05-20
申请号：JP2011240054
申请日：2011-11-01
申请人： Nippon Telegr & Teleph Corp , 日本電信電話株式会社
发明人： ARAKI AKIKO , NAKATANI TOMOHIRO
IPC分类号： G10L21/028 , H04R3/00
摘要： PROBLEM TO BE SOLVED: To provide a sound source separation technique which provides a method for efficiently estimating an arrival time difference δ and eliminates the need of overall search operation to increase the speed of estimation.SOLUTION: An arrival time difference δ, a power spectrum σof noise, a sound source spectrum s, and a sound source existence probability p(k|θ) are estimated from an observation signal x=[x,x]in a frequency domain. The estimation is performed in accordance with formula where: mis a posterior probability indicative of an expected value of attribution of the observation signal xto a sound source k; D is an interval between two microphones; c is a speed of an original signal; φ=sinc(2πfD/c) is true; ξ=[x-φ(x-s)] is true; ψand ψare phases of spectra sand ξrespectively; and δis the arrival time difference. The uncertainty of ±π which the estimated arrival time difference δincludes is corrected.
摘要翻译：要解决的问题：提供一种声源分离技术，其提供有效估计到达时间差δ的方法，并且消除了对整体搜索操作的需要以增加估计速度。
解决方案：到达时间差δ f，k ，功率谱σ f 2 ，则根据观测信号x（x）估计声源频谱s f，t，k 和声源存在概率p（k¾θ） f POS =“POST”> f，t = [x f，t，L ，x f，t， / SB>] T 。估计是按照下式进行的：其中：f，t，k 是表示观测信号的归属期望值的后验概率x f，t 到声源k; D是两个麦克风之间的间隔; c是原始信号的速度; φ f = sinc（2πfD/ c）为真; ξ f，t，k = [x f，t，R / SB>（x t，t，L -s f，t，k ）] ψ sk ，ψξk是光谱的相位s f，t，k 和ξ f，t，k 并且δ f，k 是到达时间差。校正了估计到达时间差δ f，k 的±π的不确定度。版权所有（C）2013，JPO＆INPIT

3. 发明专利

JP2012242493A Signal decomposition device, method, and program 有权
标题翻译：信号分解设备，方法和程序
公开(公告)号：JP2012242493A
公开(公告)日：2012-12-10
申请号：JP2011110463
申请日：2011-05-17
申请人： Nippon Telegr & Teleph Corp , 日本電信電話株式会社
发明人： SAWADA HIROSHI , KAMEOKA HIROKAZU , ARAKI AKIKO , UEDA SHUKO
IPC分类号： G10L21/02
摘要： PROBLEM TO BE SOLVED: To grasp a correspondence relationship between a decomposed component and a signal source when decomposing an observation signal that is formed with a mixture of signals from multiple signal sources.SOLUTION: A signal decomposition device comprises: a frequency area conversion part 12 for converting an observation signal into a time frequency component x; a preprocessing part 14 for converting the time frequency component xinto a time frequency component x'which is partially converted into a non-negative value while holding a relationship in which sensors are relative to one another; an initial value setting part 16 for setting initial values of h, t, and v; a decomposition part 18 for updating any one of vectors h, t, and v, which show transfer functions between a signal source and the sensor, so as to minimize a cost function C for approximating the sum of products of the vectors h, t, and vto the time frequency component x'after preprocessing; and a time area conversion part 20 for using a parameter when minimizing the cost function C to decompose a signal and outputs a vector hthat shows a correspondence relationship between a decomposed component and the signal source.
摘要翻译：要解决的问题：当分解由来自多个信号源的信号的混合形成的观察信号时，掌握分解的分量和信号源之间的对应关系。解决方案：信号分解装置包括：频率区域转换部分12，用于将观测信号转换为时间频率分量x ij ; 用于将时间分量x ij 转换成时间频率分量x' ij 的预处理部分14，其被部分转换为同时保持传感器相对于彼此的关系的非负值; 初始值设定部16，用于设定h ik ，t ik 的初始值，v SB SB =“POST” > KJ ; 用于更新向量h ik ，t ik 中的任何一个的分解部分18，以及v SB SB =“POST” kj ，其示出信号源和传感器之间的传递函数，以便最小化用于近似向量h ik 的乘积之和的成本函数C，时间频率分量x' ik 和v kj 预处理后; 以及时间区域转换部分20，用于当使成本函数C最小化以分解信号时使用参数，并输出显示分解成分与分解成分之间的对应关系的向量h ik 信号源。版权所有（C）2013，JPO＆INPIT

4. 发明专利

JP2012173584A Sound-source separation device, and method and program thereof 有权
标题翻译：声源分离装置及其方法和程序
公开(公告)号：JP2012173584A
公开(公告)日：2012-09-10
申请号：JP2011036559
申请日：2011-02-23
申请人： Nippon Telegr & Teleph Corp , 日本電信電話株式会社
发明人： ARAKI AKIKO , NAKATANI TOMOHIRO
IPC分类号： G10L21/02 , H04R1/40 , H04R3/00
摘要： PROBLEM TO BE SOLVED: To provide a sound-source separation technique for enabling high-accuracy, high-quality separation of a sound source in consideration of the improvement of convergence and the case of the number of sound sources, K being unknown.SOLUTION: A method includes: estimating a separation signal corresponding to each sound source from an observation signal containing one or more sound-source signals mixed with noise; converting observation signals observed by plural sound pickup means into signals of frequency domains; estimating a space parameter by using the observation signal and an occupancy ratio of a sound source representing how much each sound source contributes to the observation signal in each time frequency; estimating a model parameter including variance of sound-source signals and a spatial correlation matrix by using the observation signals and the occupancy ratios of sound sources; generating a Wiener filter based on the variance of sound-source signal and the spatial correlation matrix; estimating occupancy ratio of a sound source by using the observation signal and the spatial parameter or the model parameter; estimating a source image from the observation signals by using the Wiener filter to generate a separation signal; and converting the separation signal into a time domain signal.
摘要翻译：要解决的问题：为了提供一种声源分离技术，以便考虑到会聚的改善和声源的数量的情况下能够实现高质量的声源分离，K是未知的。解决方案：一种方法包括：从包含与噪声混合的一个或多个声源信号的观测信号估计与每个声源相对应的分离信号; 将由多个拾音装置观测到的观测信号转换为频域的信号; 通过使用所述观测信号和表示每个声源对每个时间频率对所述观测信号贡献的声源的占有率来估计空间参数; 通过使用观测信号和声源的占有率来估计包括声源信号的方差和空间相关矩阵的模型参数; 基于声源信号和空间相关矩阵的方差产生维纳滤波器; 通过使用观测信号和空间参数或模型参数来估计声源的占用率; 通过使用维纳滤波器从观测信号估计源图像以生成分离信号; 并将分离信号转换成时域信号。版权所有（C）2012，JPO＆INPIT

5. 发明专利

JP2010181467A A plurality of signals emphasizing device and method and program therefor 有权
标题翻译：多样化的信号改善设备及其方法和程序
公开(公告)号：JP2010181467A
公开(公告)日：2010-08-19
申请号：JP2009022698
申请日：2009-02-03
申请人： Nippon Telegr & Teleph Corp , 日本電信電話株式会社
发明人： ARAKI AKIKO , ISHIZUKA KENTARO , FUJIMOTO MASAKIYO , NAKATANI TOMOHIRO
IPC分类号： G10L21/02 , H04R3/00
摘要： PROBLEM TO BE SOLVED: To provide a multiple signal emphasizing device, a method and a program for improving the sound-emphasizing performance.
SOLUTION: In this plurality of signals emphasizing device, another speaker correlation matrix calculating part of this plurality of signals emphasizing device obtains another speaker correlation matrix, from the calculated value of the another speaker correlation matrix being a correlation matrix of observation signals x
b (f, t), other than a speaker k to be emphasized by inputting the observation signal x
b (f, t), cluster data C
b (t), and speaker combining information p (b) and another speaker correlation matrix of the same speaker combination in the past by checking another speaker correlation matrix holding part. A speaker to be emphasized correlation matrix calculating part calculates a speaker to be emphasized correlation matrix being a correlation matrix of the observation signal x
b (f, t) of the speaker k to be emphasized. A coefficient calculating section calculates a coefficient of SN ratio maximization beamformer related to the speaker k to be emphasized by inputting the another speaker correlation matrix and the speaker to be emphasized correlation matrix.
COPYRIGHT: (C)2010,JPO&INPIT
摘要翻译：要解决的问题：提供一种用于改善声音强调性能的多重信号强调装置，方法和程序。解决方案：在该多个信号强调装置中，该多个信号强调装置的另一个说话者相关矩阵计算部分从另一说话人相关矩阵的计算值获得另一个说话人相关矩阵，该另一说话人相关矩阵是观测信号x的相关矩阵（f，t），除了通过输入观察信号x SB SB（f，t）来强调的扬声器k以外的群集数据C＆lt; SB＆gt; b＆ / SB>（t）以及扬声器组合信息p（b）和另一扬声器组合的说话者相关矩阵，通过检查另一个说话者相关矩阵保持部分。要强调的扬声器相关矩阵计算部分计算要强调的扬声器k的观测信号x SB SB（f，t）的相关矩阵的要强调的相关矩阵。系数计算部分通过输入另一个说话者相关矩阵和要强调的说话者相关矩阵来计算要强调的扬声器k的SN比最大化波束形成器的系数。版权所有（C）2010，JPO＆INPIT

6. 发明专利

JP2004145172A Method, apparatus and program for blind signal separation, and recording medium where the program is recorded 有权
标题翻译：方法，用于盲信号分离的装置和程序，以及记录程序的记录介质
公开(公告)号：JP2004145172A
公开(公告)日：2004-05-20
申请号：JP2002312204
申请日：2002-10-28
申请人： Nippon Telegr & Teleph Corp , 日本電信電話株式会社
发明人： SAWADA HIROSHI , MUKAI MAKOTO , ARAKI AKIKO , MAKINO SHOJI
IPC分类号： G10L15/20 , G01S3/802 , G10L15/28 , G10L21/02
摘要： PROBLEM TO BE SOLVED: To highly accurately solve a permutation of blind signal separation by combining a solution of permutation by the arrival direction of a signal and a solution of permutation by the similarity of a separated signal.
SOLUTION: When the permutation is solved after (s1) performing short time Fourier transformation of an observation signal, (s2) finding separation matrixes at each frequency by independent component analysis, (s3) estimating arrival directions of signals extracted from each row of the separation matrixes at each frequency, (s4) judging whether the estimated values are reliable enough, (s5) calculating similarities of separated signals between frequencies, and (s6) finding separation matrixes at each frequency, a permutation is determined by making uniform the arrival directions of the signals at frequencies at which it is judged the estimation of the arrival directions are reliable enough, or a permutation is so determined at other frequencies that similarities of the separated signals to nearby frequencies increase.
COPYRIGHT: (C)2004,JPO
摘要翻译：要解决的问题：通过组合信号的到达方向的排列方案和通过分离的信号的相似性的排列的解决方案来高精度地解决盲信号分离的排列。解决方案：在（s1）执行观察信号的短时傅里叶变换之后解决置换时，（s2）通过独立分量分析求出每个频率处的分离矩阵，（s3）估计从每行提取的信号的到达方向（s4）判断估计值是否足够可靠，（s5）计算频率之间的分离信号的相似度，以及（s6）在每个频率下找到分离矩阵，排列通过使均匀化来确定在判断到达方向的估计的频率处的信号的到达方向是足够可靠的，或者在其他频率上如此确定在分离的信号与附近频率的相似性增加的置换。版权所有（C）2004，JPO

7. 发明专利

JP2013054258A Sound source separation device and method, and program 有权
标题翻译：声源分离装置及方法及程序
公开(公告)号：JP2013054258A
公开(公告)日：2013-03-21
申请号：JP2011193517
申请日：2011-09-06
申请人： Nippon Telegr & Teleph Corp , 日本電信電話株式会社
发明人： MERETSU SODEN , ARAKI AKIKO , KINOSHITA KEISUKE , NAKATANI TOMOHIRO , SAWADA HIROSHI
IPC分类号： G10L21/028 , G10L21/0232 , G10L21/0272
摘要： PROBLEM TO BE SOLVED: To provide a sound source separation device capable of effectively suppressing an additive noise included in an input signal.SOLUTION: In a sound source separation device, a feature vector ψ(t) featuring each time frequency bin of a multichannel observation signal y(t) is classified into N+1 components caused by each of N target sound sources and N additive noises, so that posterior probability of each of the target sound sources and additive noises is estimated by the method of maximum likelihood, and then a covariance matrix ^Rof the n-th target signal and a covariance matrix ^Rof a multichannel observation signal y(t) included in an observation signal are calculated, so that a generalized multichannel Wiener filter wfor recovering the target signal is calculated by determining unnecessary components other than the n-th target signal included in the observation signal on the basis of the covariance matrix ^Rof the n-th target signal and the covariance matrix ^Rof the multichannel observation signal y(t). Finally, an estimate ^Sof the n-th target signal is outputted on the basis of the multichannel observation signal y(t), the generalized multichannel Wiener filter w, and the posterior probability of each of the target sound sources.
摘要翻译：解决的问题：提供能够有效地抑制包含在输入信号中的附加噪声的声源分离装置。解决方案：在声源分离装置中，以多通道观测信号y（t）的每个时间频率仓为特征的特征向量ψ（t）被分类为由N个目标声源和N个目标声源中的每一个引起的N + 1个分量通过最大似然的方法估计每个目标声源和附加噪声的后验概率，然后n阶的协方差矩阵^ R xnxn 计算包含在观测信号中的多通道观测信号y（t）的第i个目标信号和协方差矩阵y R x SB SB =“POST”> yy ，使广义多通道Wiener滤波器w SB SB =“POST”> n （β）用于恢复目标信号是通过确定除观察信号中包括的第n个目标信号之外的不必要的分量来计算的第n个目标信号和协方差矩阵的协方差矩阵的基础^ R xnxn 多通道观测信号y（t）。最后，n（n）的估计值为〜 S n （β）基于多通道观测信号y（t）输出第二目标信号，广义多通道维纳滤波器w （β）< / SP>，以及每个目标声源的后验概率。版权所有（C）2013，JPO＆INPIT

8. 发明专利

JP2012173592A Sound source parameter estimation device and sound source separation device and method thereof and program therefor 有权
标题翻译：声源参数估计装置和声源分离装置及其方法及其程序
公开(公告)号：JP2012173592A
公开(公告)日：2012-09-10
申请号：JP2011036713
申请日：2011-02-23
申请人： Nippon Telegr & Teleph Corp , 日本電信電話株式会社
发明人： NAKATANI TOMOHIRO , ARAKI AKIKO , YOSHIOKA TAKUYA , FUJIMOTO MASAKIYO
IPC分类号： G10L21/02 , G10L15/14 , H04R1/40 , H04R3/00
摘要： PROBLEM TO BE SOLVED: To provide a sound source parameter estimation device capable of estimating a sound source model parameter in conjunction with a sound source parameter even when no sound source model parameter is given in advance.SOLUTION: A sound source model parameter update section updates a sound source model parameter using a sound source power feature amount, a sound source power parameter, a sound source occupancy, a preliminary probability density function of the sound source power parameter stored in a sound source model storage section, and a model of the sound source power feature amount as inputs. A sound source occupancy update section updates a sound source occupancy for each sound source using a sound source position feature amount, the sound source power feature amount, the updated sound source power parameter for each sound source, a sound source position parameter, the sound source model parameter, the preliminary probability density function of the sound source power parameter stored in the sound source model storage section, and a model of the sound source power feature amount as inputs.
摘要翻译：要解决的问题：提供一种即使在没有提供声源模型参数的情况下也能够结合声源参数来估计声源模型参数的声源参数估计装置。解决方案：声源模型参数更新部分使用声源功率特征量，声源功率参数，声源占用，存储在其中的声源功率参数的初始概率密度函数来更新声源模型参数声源模型存储部分和声源功率特征量的模型作为输入。声源占有更新部分使用声源位置特征量，声源功率特征量，每个声源的更新声源功率参数，声源位置参数，声源来更新每个声源的声源占用模型参数，存储在声源模型存储部分中的声源功率参数的初始概率密度函数，以及声源功率特征量的模型作为输入。版权所有（C）2012，JPO＆INPIT

9. 发明专利

JP2010187066A Parameter estimation apparatus, sound source separation apparatus, direction estimation apparatus, method them, and program 有权
标题翻译：参数估计装置，声源分离装置，方向估计装置，方法及程序
公开(公告)号：JP2010187066A
公开(公告)日：2010-08-26
申请号：JP2009028270
申请日：2009-02-10
申请人： Nippon Telegr & Teleph Corp , 日本電信電話株式会社
发明人： ARAKI AKIKO , NAKATANI TOMOHIRO , SAWADA HIROSHI
IPC分类号： H04R3/00 , G01S3/808 , G10L21/02
摘要： PROBLEM TO BE SOLVED: To estimate the parameter of a probability distribution for appropriately separating a sound source even when the problem of spatial aliasing arises. SOLUTION: Sound source signals from each of a plurality of sound sources are mixed, and observation signals collected in two sound collecting means are converted to a frequency domain to generate frequency observation signals (30, S102). A phase difference between the sound collecting means of the frequency observation signals is calculated (34, S104), and the parameter of a probability distribution model which is applied to the distribution of the phase difference and is frequency dependent is estimated (200, S106). COPYRIGHT: (C)2010,JPO&INPIT
摘要翻译：要解决的问题：即使出现空间混叠问题，估计用于适当地分离声源的概率分布的参数。解决方案：来自多个声源中的每一个的声源信号被混合，并且收集在两个声音采集装置中的观测信号被转换成频域以产生频率观测信号（30，S102）。计算频率观测信号的声音采集装置之间的相位差（34，S104），并且估计应用于相位差的分布并且频率相关的概率分布模型的参数（200，S106）。版权所有（C）2010，JPO＆INPIT

10. 发明专利

JP2009271183A Multiple signal sections estimation device and its method, and program and its recording medium 有权
标题翻译：多个信号部分估计装置及其方法及程序及其记录介质
公开(公告)号：JP2009271183A
公开(公告)日：2009-11-19
申请号：JP2008119717
申请日：2008-05-01
申请人： Nippon Telegr & Teleph Corp , 日本電信電話株式会社
发明人： ARAKI AKIKO , ISHIZUKA KENTARO , FUJIMOTO MASAKIYO , MAKINO SHOJI
IPC分类号： G10L21/02 , G01S3/808 , G10L11/02
摘要： PROBLEM TO BE SOLVED: To securely detect a signal section even when multiple signal sources exist at the same time.
SOLUTION: The multiple signals section estimation device includes: a frequency region conversion section; a voice existence probability estimation section; a coming direction estimation section; a coming direction probability calculation section; and a multiplication section. The voice existence probability estimation section calculates a voice existence probability in each frame. The coming direction estimation section and the coming direction probability calculation section estimate a voice coming direction probability in all frequencies in each frame. Then, the multiplication section outputs a value in which the voice existence probability is multiplied by the voice coming direction probability at the multiplication section, as an utterance probability for all sound sources. Accordingly, as existence of multiple sound sources is allowed for each frame, section detection with little deficiency is achieved.
COPYRIGHT: (C)2010,JPO&INPIT
摘要翻译：要解决的问题：即使当多个信号源同时存在时，也可以安全地检测信号部分。解决方案：多信号部分估计装置包括：频率区域转换部分; 语音存在概率估计部分; 即将到来的方向估计部分; 即将到达方向概率计算部分; 和乘法部分。语音存在概率估计部分计算每帧中的语音存在概率。前进方向估计部分和到来方向概率计算部分估计每个帧中所有频率中的语音到达方向概率。然后，乘法部在乘法部分输出语音存在概率乘以语音到达方向概率的值，作为所有声源的发声概率。因此，由于每个帧允许存在多个声源，所以实现了几乎没有的部分检测。版权所有（C）2010，JPO＆INPIT

你已经成功收藏专利！

检索式保存成功!

IPRDB

热门服务

关于我们

友情链接

联系方式