-
1.发明公开Apparatus and method for constructing multilingual acoustic model and computer readable recording medium for storing program for performing the method 审中-公开装置和用于创建多语言声学模型的方法，以及用于执行该方法用于程序存储器中的计算机可读记录介质
公开(公告)号：EP2736042A1
公开(公告)日：2014-05-28
申请号：EP13193872.2
申请日：2013-11-21
发明人： Kim, Nam-hoon , Motlicek, Petr , Garner, Philip N , Imseng, David , Lee, Jae-won , Cho, Jeong-mi
IPC分类号： G10L15/00
CPC分类号： G06F17/289 , G10L15/00
摘要： An apparatus and a method for constructing a multilingual acoustic model, and a computer readable recording medium are provided. The method for constructing a multilingual acoustic model includes dividing an input feature into a common language portion and a distinctive language portion, acquiring a tandem feature by training the divided common language portion and distinctive language portion using a neural network to estimate and remove correlation between phonemes, dividing parameters of an initial acoustic model constructed using the tandem feature into common language parameters and distinctive language parameters, adapting the common language parameters using data of a training language, adapting the distinctive language parameters using data of a target language, and constructing an acoustic model for the target language using the adapted common language parameters and the adapted distinctive language parameters.
摘要翻译： 本发明提供一种装置和用于构建多语言声学模型的方法，以及一种计算机可读记录介质。 用于构建多语言声学模型的方法包括：输入特征的分割到一个共同的语言部分和一个独特的语言部分，通过使用神经网络来估计和去除音素之间的相关性训练划分的共同语言部分和区别语言部分获取串联特征 中，初始声学模型的分割参数使用串联特征为共同语言参数和区别语言参数，适应使用锻炼语言的数据的共同语言的参数，适应使用目标语言的数据区别语言参数构造，并且声学构建 模型使用angepasst共同语言参数和angepasst的区别语言参数目标语言。
-
公开(公告)号：EP2434532A1
公开(公告)日：2012-03-28
申请号：EP11183069.1
申请日：2011-09-28
发明人： Tak, Young-jo , Lee, Jae-won , Park, Young-soo , Kim, Jun-youn
IPC分类号： H01L21/02
CPC分类号： H01L21/0254 , H01L21/02378 , H01L21/02381 , H01L21/02458 , H01L21/02502 , H01L21/02505 , H01L21/02507 , H01L21/0259 , H01L21/02642 , H01L21/02664 , H01L33/007 , H01L33/12 , H01L33/32
摘要： A semiconductor device includes a substrate, a buffer layer on the substrate, and a plurality of nitride semiconductor layers on the buffer layer. The semiconductor device further includes at least one masking layer and at least one inter layer between the plurality of nitride semiconductor layers. The at least one inter layer is on the at least one masking layer.
摘要翻译： 一种半导体器件包括衬底，衬底上的缓冲层以及缓冲层上的多个氮化物半导体层。 该半导体器件还包括至少一个掩模层和在该多个氮化物半导体层之间的至少一个中间层。 该至少一个中间层位于至少一个掩模层上。
-
3.发明授权Method of setting optimum-partitioned classified neural network and method and apparatus for automatic labeling using optimum-partitioned classified neural network 有权一种用于使用最佳分区，klassifierten神经网络形成的最佳分配，klassifierten神经网络，和方法和装置用于自动标记处理
公开(公告)号：EP1453037B1
公开(公告)日：2010-06-09
申请号：EP04251145.1
申请日：2004-02-27
发明人： Choo, Ki-hun , Lee, Jae-won , Kim, Jeong-su , Lee, Ki-seung
IPC分类号： G10L15/04
CPC分类号： G10L15/04
摘要： A method of setting an optimum-partitioned classified neural network and a method and apparatus for automatic labeling using an optimum-partitioned classified neural network are provided. The method of automatic labeling using an optimum-partitioned classified neural network comprises (a) searching for neural networks having minimum errors with respect to a number of L phoneme combinations from a number of K neural network combinations generated at an initial stage or updated, updating weights during learning of the K neural networks by K phoneme combination groups searched with the same neural networks, and composing an optimum-partitioned classified neural network combination using the K neural networks of which a total error sum has converged; and (b) tuning a phoneme boundary of a first label file by using the phoneme combination group classification result and the optimum-partitioned classified neural network combination, and generating a final label file reflecting the tuning result.
-
公开(公告)号：EP2979637B1
公开(公告)日：2018-09-12
申请号：EP15179033.4
申请日：2015-07-30
CPC分类号： A61B6/4429 , A61B6/4233 , A61B6/4283 , A61B6/4291 , A61B6/4464 , A61B6/547 , A61B6/56 , G03B42/04
摘要： An X-ray imaging apparatus includes an X-ray detecting device, an accommodation unit configured to accommodate the X-ray detecting device, and a bucky tray configured to be accommodated in and discharged from the accommodation unit. The bucky tray includes a base and a first plate on which the X-ray detecting device is mounted and which is configured to rotate with respect to the base.
-
5.发明公开Method of recognizing speech and electronic device thereof 审中-公开Verfahren zur Spracherkennung und elektronische Vorrichtungdafür
公开(公告)号：EP2685452A1
公开(公告)日：2014-01-15
申请号：EP13174723.0
申请日：2013-07-02
发明人： Lee, Jae-won , Yook, Dong-suk , Lim, Hyeon-taek , Kim, Tae-yoon
CPC分类号： G10L15/18 , G10L15/05 , G10L15/142
摘要： A method of recognizing a speech and an electronic device thereof are provided. The method includes: segmenting a speech signal into a plurality of sections at preset time intervals; performing a phoneme recognition with respect to one of the plurality of sections of the speech signal by using a first acoustic model; extracting a candidate word of the one of the plurality of sections of the speech signal by using the phoneme recognition result; and performing a speech recognition with respect to the one the plurality of sections the speech signal by using the candidate word.
摘要翻译： 提供了一种识别语音及其电子设备的方法。 该方法包括：以预设的时间间隔将语音信号分割成多个部分; 通过使用第一声学模型对语音信号的多个部分之一执行音素识别; 通过使用音素识别结果提取语音信号的多个部分中的一个部分的候选词; 以及通过使用所述候选词对所述语音信号的所述多个部分中的所述部分执行语音识别。
-
6.发明授权
-
7.发明公开Method of setting optimum-partitioned classified neural network and method and apparatus for automatic labeling using optimum-partitioned classified neural network 有权一种用于使用最佳分区，klassifierten神经网络形成的最佳分配，klassifierten神经网络，和方法和装置用于自动标记处理
公开(公告)号：EP1453037A3
公开(公告)日：2006-05-17
申请号：EP04251145.1
申请日：2004-02-27
发明人： Choo, Ki-hun , Lee, Jae-won , Kim, Jeong-su , Lee, Ki-seung
IPC分类号： G10L15/04
CPC分类号： G10L15/04
摘要： A method of setting an optimum-partitioned classified neural network and a method and apparatus for automatic labeling using an optimum-partitioned classified neural network are provided. The method of automatic labeling using an optimum-partitioned classified neural network comprises (a) searching for neural networks having minimum errors with respect to a number of L phoneme combinations from a number of K neural network combinations generated at an initial stage or updated, updating weights during learning of the K neural networks by K phoneme combination groups searched with the same neural networks, and composing an optimum-partitioned classified neural network combination using the K neural networks of which a total error sum has converged; and (b) tuning a phoneme boundary of a first label file by using the phoneme combination group classification result and the optimum-partitioned classified neural network combination, and generating a final label file reflecting the tuning result.
-
8.发明公开Method of setting optimum-partitioned classified neural network and method and apparatus for automatic labeling using optimum-partitioned classified neural network 有权一种用于使用最佳分区，klassifierten神经网络形成的最佳分配，klassifierten神经网络，和方法和装置用于自动标记处理
公开(公告)号：EP1453037A2
公开(公告)日：2004-09-01
申请号：EP04251145.1
申请日：2004-02-27
发明人： Choo, Ki-hun , Lee, Jae-won , Kim, Jeong-su , Lee, Ki-seung
IPC分类号： G10L15/04
CPC分类号： G10L15/04
摘要： A method of setting an optimum-partitioned classified neural network and a method and apparatus for automatic labeling using an optimum-partitioned classified neural network are provided. The method of automatic labeling using an optimum-partitioned classified neural network comprises (a) searching for neural networks having minimum errors with respect to a number of L phoneme combinations from a number of K neural network combinations generated at an initial stage or updated, updating weights during learning of the K neural networks by K phoneme combination groups searched with the same neural networks, and composing an optimum-partitioned classified neural network combination using the K neural networks of which a total error sum has converged; and (b) tuning a phoneme boundary of a first label file by using the phoneme combination group classification result and the optimum-partitioned classified neural network combination, and generating a final label file reflecting the tuning result.
摘要翻译： 设置在最佳分区分类的神经网络，并用于使用在最佳分区分类的神经网络的自动贴标签的方法和装置的设置的方法。 自动贴标签的使用在最佳分区归类神经网络包括方法（a）搜索具有相对于误差最小的神经网络的数量L音素组合的从许多在初始阶段或更新生成的K个神经网络的组合，更新 用K音素组合组学习K个神经网络的权重过程中搜索用相同的神经网络，并在构成最佳分区分类使用的是哪个的K个神经网络总误差总和已经收敛神经网络组合; 和（b）通过使用音素组组合的分类结果和最佳分区分类的神经网络组合，并且生成最终标签文件反映调谐结果调谐的第一标签文件的音素边界。
-
公开(公告)号：EP2287924B1
公开(公告)日：2019-02-27
申请号：EP10172937.4
申请日：2010-08-16
发明人： Kim, Jun-youn , Jeong, Hyung-su , Lee, Jae-won , Tak, Young-jo , Hong, Hyun-gi
-
10.发明公开Semiconductor buffer structure, semiconductor device including the same, and method of manufacturing semiconductor device using semiconductor buffer structure 审中-公开半导体缓冲结构，使半导体器件和工艺用于使用所述半导体缓冲结构的半导体器件制造
公开(公告)号：EP2696365A2
公开(公告)日：2014-02-12
申请号：EP13179953.8
申请日：2013-08-09
发明人： Tak, Young-jo , Kim, Jae-kyun , Kim, Joo-sung , Kim, Jun-youn , Lee, Jae-won , Choi, Hyo-ji
IPC分类号： H01L21/02
CPC分类号： H01L33/12 , H01L21/02381 , H01L21/02458 , H01L21/02505 , H01L21/02507 , H01L21/0251 , H01L21/0254 , H01L33/0066 , H01L33/0079
摘要： Provided is a method of manufacturing a semiconductor device. The method includes preparing a silicon substrate, forming a buffer layer on the silicon substrate, and forming a nitride semiconductor layer on the buffer layer. The buffer layer includes a first layer, a second layer, and a third layer. The first layer includes Al x In y Ga 1-x-y N (0≤x≤1, 0≤y≤1, 0≤x+y≤1) and has a lattice constant LP1 that is smaller than a lattice constant LP0 of the silicon substrate. The second layer is formed on the first layer, includes Al x In y Ga 1-x-y N (0≤x x In y Ga 1-x-y N (0≤x
摘要翻译： 提供了一种制造半导体器件的方法。 该方法包括制备一硅衬底，形成在硅衬底上的缓冲层，和在所述缓冲层上的氮化物半导体层。 该缓冲层包括第一层，第二层和第三层。 所述第一层包含Al X的In y镓1-XY N（0‰¤x‰¤1，0‰¤y‰¤1，0‰¤x+ Y‰¤1），并具有的晶格常数LP1确实是小于 所述硅衬底的晶格常数LP0。 第二层在第一层上形成的，包含Al X的In y镓1-XY N（0‰¤x<1,0‰¤y<1,0‰¤x+ Y <1），并且具有的晶格常数 LP2确实比LP0大于LP1小。 第三层是在第二层上形成的，包含Al X的In y镓1-XY N（0‰¤x<1,0‰¤y<1,0‰¤x+ Y <1），并且具有的晶格常数 LP3确实比LP2小。
-
-
-
-
-
-
-
-
-
搜索结果
25 条记录 (耗时 0.015 秒)
