Advanced missing feature theory with fast score calculation for noise robust speaker identification

被引:3
作者
Jung, J. [1 ]
Kim, K. [1 ]
Kim, M. Y. [1 ]
机构
[1] Sejong Univ, Dept Informat & Commun Engn, Biometr Engn Res Ctr, Seoul, South Korea
关键词
SPEECH RECOGNITION; MODELS;
D O I
10.1049/el.2010.0368
中图分类号
TM [电工技术]; TN [电子技术、通信技术];
学科分类号
0808 ; 0809 ;
摘要
Background noise robustness is one of the major concerns in speaker identification systems. Extended missing feature theory (EMFT) has been applied for this application by ignoring the unreliable parts of the input feature. Proposed is the advanced missing feature theory (AMFT) to improve the identification rate of EMFT by considering the cross-terms in a feature vector. To reduce the computational complexity of AMFT, also proposed is the fast-AMFT (F-AMFT) by applying the recursive score-calculation algorithm. Compared with EMFT, F-AMFT requires around 400 times lower complexity and gives significantly better identification rate even for non-stationary background noise environments.
引用
收藏
页码:1027 / 1028
页数:2
相关论文
共 8 条
[1]  
Bourlard H., 1996, P ICSLP PHIL US OCT
[2]  
HERMANSKY H, 1996, P ICSLP PHIL US OCT
[3]   Noise compensation for speech recognition with arbitrary additive noise [J].
Ming, J .
IEEE TRANSACTIONS ON AUDIO SPEECH AND LANGUAGE PROCESSING, 2006, 14 (03) :833-844
[4]   Robust speech recognition using probabilistic union models [J].
Ming, J ;
Jancovic, P ;
Smith, FJ .
IEEE TRANSACTIONS ON SPEECH AND AUDIO PROCESSING, 2002, 10 (06) :403-414
[5]   Robust speaker recognition in noisy conditions [J].
Ming, Ji ;
Hazen, Timothy J. ;
Glass, James R. ;
Reynolds, Douglas A. .
IEEE TRANSACTIONS ON AUDIO SPEECH AND LANGUAGE PROCESSING, 2007, 15 (05) :1711-1723
[6]  
Pullella D., 2008, ROBUST SPEAKER IDENT
[7]   ROBUST TEXT-INDEPENDENT SPEAKER IDENTIFICATION USING GAUSSIAN MIXTURE SPEAKER MODELS [J].
REYNOLDS, DA ;
ROSE, RC .
IEEE TRANSACTIONS ON SPEECH AND AUDIO PROCESSING, 1995, 3 (01) :72-83
[8]  
Vizinho A., 1999, P EUR, V99, P2407