%0 Journal Article
%T A Robust Front-end for Speech Recognition Based on Computational Auditory Scene Analysis and Speaker Model<br>基于计算听觉场景分析和语者模型信息的语音识别鲁棒前端研究
%A GUAN Yong LI Peng LIU Wen-Ju XU Bo National Laboratory of Pattern Recognition
%A Institute of Automation
%A Chinese Academy of Sciences
%A Beijing Nokia Research Center
%A Beijing Digital Content Technology Research Center
%A Beijing
%A <br>关勇
%A 李鹏
%A 刘文举
%A 徐波
%J 自动化学报
%D 2009
%I 
%X Conventional noise robust speech recognition system does not work well when human speech is presented in the background. In this paper, a computational auditory scene analysis (CASA) and speaker model based speech segregation system is proposed to solve this problem. By utilizing speaker model and factorial-max vector quantization (MAXVQ) to estimate real-value masks in CASA framework, a robust front-end for speech recognition is constructed. Evaluations on speech separation challenge (SSC) showed that the proposed system won 15.68% improvement over the baseline system. The results of evaluation also proved the validity of the multi-speaker recognition and the real-value mask estimation module.
%K Computational auditory scene analysis (CASA)
%K speech segregation
%K robust speech recognition
%K factorial-max vector quantization (MAXVQ)
%K speaker recognition<br>计算听觉场景分析
%K 语音分离
%K 鲁棒语音识别
%K 因子最大矢量量化
%K 语者识别
%U http://www.alljournals.cn/get_abstract_url.aspx?pcid=5B3AB970F71A803DEACDC0559115BFCF0A068CD97DD29835&cid=8240383F08CE46C8B05036380D75B607&jid=E76622685B64B2AA896A7F777B64EB3A&aid=EB9ACF64409212549DCBCECBD2C4E641&yid=DE12191FBD62783C&vid=6209D9E8050195F5&iid=E158A972A605785F&sid=FED44C0135DC1D9C&eid=1FF3CD54EFC256A1&journal_id=0254-4156&journal_name=自动化学报&referenced_num=1&reference_num=0