发明授权
US07444282B2 Method of setting optimum-partitioned classified neural network and method and apparatus for automatic labeling using optimum-partitioned classified neural network
有权
设置最优分割分类神经网络的方法以及使用最优分割分类神经网络进行自动标注的方法和装置
- 专利标题: Method of setting optimum-partitioned classified neural network and method and apparatus for automatic labeling using optimum-partitioned classified neural network
- 专利标题(中): 设置最优分割分类神经网络的方法以及使用最优分割分类神经网络进行自动标注的方法和装置
-
申请号: US10788301申请日: 2004-03-01
-
公开(公告)号: US07444282B2公开(公告)日: 2008-10-28
- 发明人: Ki-hyun Choo , Jeong-su Kim , Jae-won Lee , Ki-seung Lee
- 申请人: Ki-hyun Choo , Jeong-su Kim , Jae-won Lee , Ki-seung Lee
- 申请人地址: KR Suwon-Si
- 专利权人: Samsung Electronics Co., Ltd.
- 当前专利权人: Samsung Electronics Co., Ltd.
- 当前专利权人地址: KR Suwon-Si
- 代理机构: Staas & Halsey LLP
- 优先权: KR10-2003-0012700 20030228
- 主分类号: G10L11/00
- IPC分类号: G10L11/00 ; G10L21/00 ; G10L11/06 ; G10L21/06 ; G10L15/00 ; G10L15/16 ; G10L15/06 ; G10L13/00 ; G06E1/00 ; G06E3/00 ; G06F15/18 ; G06G7/00 ; G06N3/08
摘要:
A method of automatic labeling using an optimum-partitioned classified neural network includes 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 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/授权文献
信息查询