公开(公告)号：US08976941B2

公开(公告)日：2015-03-10

申请号：US11928665

申请日：2007-10-30

申请人： Hyun-Soo Kim ,  Myeong-Gi Jeong ,  Young-Hee Park

发明人： Hyun-Soo Kim ,  Myeong-Gi Jeong ,  Young-Hee Park

IPC分类号： H04M1/64 ,  G10L15/01 ,  G10L25/00

CPC分类号： G10L15/01 ,  G10L25/00 ,  G10L2015/225

摘要： Provided are an apparatus and method for reporting speech recognition failures. The method includes detecting pure speech data from input speech data and outputting the detected pure speech data; checking at least one speech recognition failure for the pure speech data; and ascertaining speech recognition failure reasons from a check-result for the speech recognition failures and outputting the ascertained speech recognition failure reasons.

摘要翻译： 提供了用于报告语音识别失败的装置和方法。 该方法包括从输入语音数据检测纯语音数据并输出检测到的纯语音数据; 检查纯语音数据的至少一个语音识别失败; 并从语音识别失败的检查结果中确定语音识别失败原因并输出确定的语音识别失败原因。

公开(公告)号：US08135220B2

公开(公告)日：2012-03-13

申请号：US12115250

申请日：2008-05-05

申请人： Hyun-Soo Kim ,  Je-Han Yoon ,  Hyun-Sik Shim ,  Young-Hee Park ,  Do-Hyung Kim ,  Woo-Han Yun

发明人： Hyun-Soo Kim ,  Je-Han Yoon ,  Hyun-Sik Shim ,  Young-Hee Park ,  Do-Hyung Kim ,  Woo-Han Yun

IPC分类号： G06K9/62 ,  G06K9/00

CPC分类号： G06K9/00295

摘要： A face recognition system based on adaptive learning includes a specific person detection and tracking unit for detecting and tracking a specific person from a moving image. A facial feature extraction unit extracts a plurality of facial feature vectors from the detected and tracked specific person. A face recognition unit searches for a given registration model by comparing the extracted facial feature vectors with facial feature vectors of the registration models previously stored in a user registration model database. A learning target selection unit selects a facial feature vector to be added to a record of the given registration model from among the extracted facial feature vectors. A registration model learning unit adds and updates the selected facial feature vector to the record of the given registration model.

摘要翻译： 基于自适应学习的面部识别系统包括用于从运动图像中检测和跟踪特定人物的特定人物检测和跟踪单元。 面部特征提取单元从检测到和跟踪的特定人物中提取多个面部特征向量。 面部识别单元通过将提取的面部特征向量与先前存储在用户登记模型数据库中的登记模型的面部特征向量进行比较来搜索给定的登记模型。 学习目标选择单元从所提取的面部特征向量中选择要添加到给定登记模型的记录的面部特征向量。 注册模型学习单元将所选择的面部特征向量添加并更新到给定注册模型的记录。

公开(公告)号：US20070201639A1

公开(公告)日：2007-08-30

申请号：US11705802

申请日：2007-02-13

申请人： Young-Hee Park ,  Hyun-Soo Kim ,  Hyun-Sik Shim

发明人： Young-Hee Park ,  Hyun-Soo Kim ,  Hyun-Sik Shim

IPC分类号： H04M11/00

CPC分类号： G10L15/30

摘要： Provided is a system and method for controlling voice detection of a network terminal. The system includes the network terminal for, if detection of a voice signal is requested, detecting voice by receiving and setting a voice detection setting value corresponding to a predetermined service and generating a trigger signal for the voice detection according to the voice detection setting value corresponding to the service; and a server for determining the service of the network terminal and transmitting the voice detection setting value corresponding to the service to the network terminal. Accordingly, by controlling to commence voice detection according to a service, voice detection optimized to a relevant service can commence.

摘要翻译： 提供一种用于控制网络终端的语音检测的系统和方法。 该系统包括网络终端，如果要求检测到语音信号，则通过接收并设置与预定服务相对应的语音检测设置值来检测语音，并根据对应的语音检测设置值产生用于语音检测的触发信号 服务; 以及服务器，用于确定网络终端的服务，并将对应于该服务的语音检测设置值发送到网络终端。 因此，通过控制根据业务开始语音检测，可以开始优化到相关服务的语音检测。

公开(公告)号：US20080273766A1

公开(公告)日：2008-11-06

申请号：US12115250

申请日：2008-05-05

申请人： Hyun-Soo Kim ,  Je-Han Yoon ,  Hyun-Sik Shim ,  Young-Hee Park ,  Do-Hyung Kim ,  Woo-Han Yun

发明人： Hyun-Soo Kim ,  Je-Han Yoon ,  Hyun-Sik Shim ,  Young-Hee Park ,  Do-Hyung Kim ,  Woo-Han Yun

IPC分类号： G06K9/00

CPC分类号： G06K9/00295

摘要： A face recognition system based on adaptive learning includes a specific person detection and tracking unit for detecting and tracking a specific person from a moving image. A facial feature extraction unit extracts a plurality of facial feature vectors from the detected and tracked specific person. A face recognition unit searches for a given registration model by comparing the extracted facial feature vectors with facial feature vectors of the registration models previously stored in a user registration model database. A learning target selection unit selects a facial feature vector to be added to a record of the given registration model from among the extracted facial feature vectors. A registration model learning unit adds and updates the selected facial feature vector to the record of the given registration model.

摘要翻译： 基于自适应学习的面部识别系统包括用于从运动图像中检测和跟踪特定人物的特定人物检测和跟踪单元。 面部特征提取单元从检测到和跟踪的特定人物中提取多个面部特征向量。 面部识别单元通过将提取的面部特征向量与先前存储在用户登记模型数据库中的登记模型的面部特征向量进行比较来搜索给定的登记模型。 学习目标选择单元从所提取的面部特征向量中选择要添加到给定登记模型的记录的面部特征向量。 注册模型学习单元将所选择的面部特征向量添加并更新到给定注册模型的记录。

公开(公告)号：US20080140424A1

公开(公告)日：2008-06-12

申请号：US11954717

申请日：2007-12-12

申请人： Hyun-Soo Kim ,  Hyun-Sik Shim ,  Young-Hee Park ,  Je-Han Yoon ,  Jong-Gyu Ham

发明人： Hyun-Soo Kim ,  Hyun-Sik Shim ,  Young-Hee Park ,  Je-Han Yoon ,  Jong-Gyu Ham

IPC分类号： H04N7/173

CPC分类号： H04N5/772 ,  H04N5/23219 ,  H04N5/775 ,  H04N9/7921 ,  H04N9/8063 ,  H04N2101/00 ,  H04N2201/0084 ,  H04N2201/3264 ,  H04N2201/3266

摘要： A auto-recording method is disclosed for auto-recording further to user request, via generating user image and voice data, extracting feature points from the image data according to pre-defined user recognition and following by considering the user as an object of following according to extracted feature points, determining whether the image and voice data satisfy a recording reference needed to perform recording. If determined that the image and voice data satisfy the recording reference, editing the image and voice data in a pre-set edit form and generating and storing at least one of recording image and recording voice data.

摘要翻译： 公开了一种自动记录方法，用于通过生成用户图像和语音数据进一步自动记录用户请求，根据预定义的用户识别从图像数据中提取特征点，并且通过将用户作为跟随对象来考虑， 提取特征点，确定图像和语音数据是否满足执行记录所需的记录参考。 如果确定图像和语音数据满足记录参考，则以预设编辑形式编辑图像和语音数据，并且生成并存储记录图像和记录语音数据中的至少一个。

公开(公告)号：US07890334B2

公开(公告)日：2011-02-15

申请号：US11705802

申请日：2007-02-13

申请人： Young-Hee Park ,  Hyun-Soo Kim ,  Hyun-Sik Shim

发明人： Young-Hee Park ,  Hyun-Soo Kim ,  Hyun-Sik Shim

IPC分类号： G10L15/22

CPC分类号： G10L15/30

摘要： Provided is a system and method for controlling voice detection of a network terminal. The system includes the network terminal for, if detection of a voice signal is requested, detecting voice by receiving and setting a voice detection setting value corresponding to a predetermined service and generating a trigger signal for the voice detection according to the voice detection setting value corresponding to the service; and a server for determining the service of the network terminal and transmitting the voice detection setting value corresponding to the service to the network terminal. Accordingly, by controlling to commence voice detection according to a service, voice detection optimized to a relevant service can commence.

摘要翻译： 提供一种用于控制网络终端的语音检测的系统和方法。 该系统包括网络终端，如果要求检测到语音信号，则通过接收并设置与预定服务相对应的语音检测设置值来检测语音，并根据对应的语音检测设置值产生用于语音检测的触发信号 服务; 以及服务器，用于确定网络终端的服务，并将对应于该服务的语音检测设置值发送到网络终端。 因此，通过控制根据业务开始语音检测，可以开始优化到相关服务的语音检测。

公开(公告)号：US08849662B2

公开(公告)日：2014-09-30

申请号：US11646911

申请日：2006-12-28

申请人： Hyun-Soo Kim

发明人： Hyun-Soo Kim

IPC分类号： G10L15/02 ,  G10L15/00 ,  G10L15/06 ,  G10L17/00 ,  G10L15/04

CPC分类号： G10L15/04 ,  G10L2015/025

摘要： A method and a system for segmenting phonemes from voice signals. A method for accurately segmenting phonemes, in which a histogram showing a peak distribution corresponding to an order is formed by using a high order concept, and a boundary indicating a starting point and an ending point of each phoneme is determined by calculating a peak statistic based on the histogram. The phoneme segmentation method can remarkably reduce an amount of calculation, and has an advantage of being applied to sound signal systems which perform sound coding, sound recognition, sound synthesizing, sound reinforcement, etc.

摘要翻译： 一种从语音信号中分割音素的方法和系统。 一种用于准确分割音素的方法，其中通过使用高阶概念形成表示与顺序相对应的峰值分布的直方图，并且通过计算基于峰值统计量来确定表示每个音素的起点和终点的边界 在直方图上。 音素分割方法可以显着减少计算量，并且具有应用于执行声音编码，声音识别，声音合成，扩音等的声音信号系统的优点。

公开(公告)号：US08577677B2

公开(公告)日：2013-11-05

申请号：US12460473

申请日：2009-07-20

申请人： Hyun-Soo Kim ,  Hanseok Ko ,  Jounghoon Beh ,  Taekjin Lee

发明人： Hyun-Soo Kim ,  Hanseok Ko ,  Jounghoon Beh ,  Taekjin Lee

IPC分类号： G10L21/02

CPC分类号： G10L21/0272 ,  G10L2021/02166

摘要： A system and method for sound source separation. The system and method use a beamforming technique. The sound source separation system includes a windowing processor; a DFT transformer; a transfer function estimator; and a noise estimator. The system also includes a voice signal extractor that cancels individual voice signals, except an individual voice signal that is desired to be extracted among individual voice signals, from the integrated voice signals. The system further includes a voice signal detector that cancels a noise part provided through the noise estimator from a transfer function of an individual voice signal which is desired to be detected and extracts a noise-canceled individual voice signal. Even when two or more sound sources are simultaneously input, the sound sources can be separated from each other and separately stored and managed, or an initial sound source can be stored and managed.

摘要翻译： 一种声源分离的系统和方法。 该系统和方法使用波束成形技术。 声源分离系统包括开窗处理器; DFT变压器; 传递函数估计器; 和噪声估计器。 该系统还包括语音信号提取器，除了从各个语音信号中提取的单个语音信号之外，还可以从集成语音信号中消除各个语音信号。 该系统还包括语音信号检测器，该语音信号检测器根据期望被检测的单独语音信号的传递函数来消除通过噪声估计器提供的噪声部分，并提取噪声消除的个体语音信号。 即使当同时输入两个或更多个声源时，声源可以彼此分开并分开存储和管理，或者可以存储和管理初始声源。

公开(公告)号：US08331690B2

公开(公告)日：2012-12-11

申请号：US12726762

申请日：2010-03-18

申请人： Sang-Wook Oh ,  Seong-Taek Hwang ,  Hyun-Soo Kim ,  Sang-Ho Kim ,  Ho-Seok An ,  Mu-Sik Kwon ,  Ji-Hoon Kim ,  Dong-Chang Lee

发明人： Sang-Wook Oh ,  Seong-Taek Hwang ,  Hyun-Soo Kim ,  Sang-Ho Kim ,  Ho-Seok An ,  Mu-Sik Kwon ,  Ji-Hoon Kim ,  Dong-Chang Lee

IPC分类号： G06K9/48 ,  G06K9/36 ,  G06K9/32

CPC分类号： G06K9/32 ,  G06K2009/2045 ,  H04N1/3876 ,  H04N5/232 ,  H04N5/23238

摘要： A method is provided for creating a panorama. The method includes photographing a plurality of images having same backgrounds and different forms of a subject, determining a size and a position of a reference region for creating a panorama using the images, extracting a target region within the reference region from each of the images, detecting same portions in adjacent target regions, and creating a panorama by combining the adjacent target regions on the basis of the same portions.

摘要翻译： 提供了一种用于创建全景图的方法。 该方法包括拍摄具有相同背景和不同形式的对象的多个图像，使用该图像确定用于创建全景的参考区域的大小和位置，从每个图像提取参考区域内的目标区域， 检测相邻目标区域中的相同部分，以及通过基于相同部分组合相邻目标区域来创建全景图。

公开(公告)号：US08135586B2

公开(公告)日：2012-03-13

申请号：US12053144

申请日：2008-03-21

申请人： Hyun-Soo Kim ,  Hanseok Ko ,  Sung-Joo Ahn ,  Jounghoon Beh ,  Hyun-Jin Yoon

发明人： Hyun-Soo Kim ,  Hanseok Ko ,  Sung-Joo Ahn ,  Jounghoon Beh ,  Hyun-Jin Yoon

IPC分类号： G10L19/10

CPC分类号： G10L21/0208

摘要： Disclosed is a method and an apparatus for estimating noise included in a sound signal during sound signal processing. The method includes estimating harmonics components in a frame of an input sound signal; using the estimated harmonics components, computing a Voice Presence Probability (VPP) on the frame of the input sound signal; determining a weight of an equation necessary to estimate a noise spectrum, depending on the computed VPP; and using the determined weight and the equation necessary to estimate a noise spectrum, estimating the noise spectrum, and updating the noise spectrum.

摘要翻译： 公开了一种用于在声音信号处理期间估计包括在声音信号中的噪声的方法和装置。 该方法包括估计输入声音信号的帧中的谐波分量; 使用估计的谐波分量，在输入声音信号的帧上计算声音概率（VPP）; 根据所计算的VPP确定估计噪声谱所必需的方程的权重; 并且使用所确定的权重和所需的等式来估计噪声谱，估计噪声谱和更新噪声谱。