作者其他论文
文献详情
A multi-stream bimodal continuous speech recognition system using datasieve based features
文献类型:会议
作者:Xie, Lei[1]  Ravyse, Ilse[2]  Jiang, Dong-Mei[3]  Zhao, Rong-Chun[4]  Sahli, Hichem[5]  Verhelst, Werver[6]  Cornelis, Jan[7]  
机构:[1]Dept. Comp. Science and Engineering, NW. Polytechnical University, Xi'an 710072, China
[2]Dept. ETRO, Vrije Universiteit Brussel, Pleinlaan 2, 1050 Brussels, Belgium
[3]Dept. Comp. Science and Engineering, NW. Polytechnical University, Xi'an 710072, China
[4]Dept. Comp. Science and Engineering, NW. Polytechnical University, Xi'an 710072, China
[5]Dept. ETRO, Vrije Universiteit Brussel, Pleinlaan 2, 1050 Brussels, Belgium
[6]Dept. ETRO, Vrije Universiteit Brussel, Pleinlaan 2, 1050 Brussels, Belgium
[7]Dept. ETRO, Vrije Universiteit Brussel, Pleinlaan 2, 1050 Brussels, Belgium
年:2003
通讯作者:Xie, L.(xielei@263.net)
会议名称:2003 International Conference on Machine Learning and Cybernetics
页码范围:2287-2290
会议地点:Xi'an, China
会议开始日期:2003-11-02
会议结束日期:2003-11-05
收录情况:EI(2004128072262)  
人气指数:1880
浏览次数:1859
语言:外文
摘要:
This paper presents an audio visual bimodal continuous speech recognition system. The visual feature extraction of the mouth movements uses the number of granules obtained by applying a datasieve. Multi-stream HMMs are introduced for combining audio and visual modalities using time synchronous audio visual features. Experimental results show that the recognition system provided by this paper is suitable for continuous speech recognition tasks in noisy environments, and the datasieve based visual ...More
0
评论(0 条评论)
登录