LEADER 04619oam 2200505 450 001 9910825222703321 005 20190911100031.0 010 $a0-12-800253-0 035 $a(OCoLC)870677231 035 $a(MiFhGG)GVRL8CUI 035 $a(EXLCZ)992550000001182738 100 $a20140529d2014 uy 0 101 0 $aeng 135 $aurun|---uuuua 181 $ctxt 182 $cc 183 $acr 200 10$aSpeech enhancement $ea signal subspace perspective /$fJacob Benesty [and three others] 205 $a1st ed. 210 1$aOxford :$cAcademic Press,$d2014. 215 $a1 online resource (vi, 135 pages) $ccolor illustrations 225 0 $aGale eBooks 300 $aDescription based upon print version of record. 311 $a0-12-800139-9 311 $a1-306-31513-1 320 $aIncludes bibliographical references at the end of each chapters and index. 327 $aHalf Title; Title Page; Copyright; Contents; 1 Introduction; 1.1 History and Applications of Subspace Methods; 1.2 Speech Enhancement from a Signal Subspace Perspective; 1.3 Scope and Organization of the Work; References; 2 General Concept with the Diagonalization of the Speech Correlation Matrix; 2.1 Signal Model and Problem Formulation; 2.2 Linear Filtering with a Rectangular Matrix; 2.3 Performance Measures; 2.3.1 Noise Reduction; 2.3.2 Speech Distortion; 2.3.3 MSE Criterion; 2.4 Optimal Rectangular Filtering Matrices; 2.4.1 Maximum SNR; 2.4.2 Wiener; 2.4.3 MVDR; 2.4.4 Tradeoff 327 $a2.4.5 LCMVReferences; 3 General Concept with the Joint Diagonalization of the Speech and Noise Correlation Matrices; 3.1 Signal Model and Problem Formulation; 3.2 Linear Filtering with a Rectangular Matrix; 3.3 Performance Measures; 3.3.1 Noise Reduction; 3.3.2 Speech Distortion; 3.3.3 MSE Criterion; 3.4 Optimal Rectangular Filtering Matrices; 3.4.1 Maximum SNR; 3.4.2 Wiener; 3.4.3 MVDR; 3.4.4 Tradeoff; 3.5 Another Signal Model; References; 4 Single-Channel Speech Enhancement in the Time Domain; 4.1 Signal Model and Problem Formulation; 4.2 Linear Filtering with a Rectangular Matrix 327 $a4.3 Performance Measures4.4 Optimal Rectangular Filtering Matrices; 4.5 Single-Channel Noise Reduction Revisited; 4.5.1 Orthogonal Decomposition; 4.5.2 Linear Filtering with a Rectangular Matrix; 4.5.3 Performance Measures; 4.5.4 Optimal Rectangular Filtering Matrices; References; 5 Multichannel Speech Enhancement in the Time Domain; 5.1 Signal Model and Problem Formulation; 5.2 Linear Filtering with a Rectangular Matrix; 5.3 Performance Measures; 5.3.1 Noise Reduction; 5.3.2 Speech Distortion; 5.3.3 MSE Criterion; 5.4 Optimal Rectangular Filtering Matrices; 5.4.1 Maximum SNR; 5.4.2 Wiener 327 $a5.4.3 MVDR5.4.4 Tradeoff; 5.4.5 LCMV; References; 6 Multichannel Speech Enhancement in the Frequency Domain; 6.1 Signal Model and Problem Formulation; 6.2 Linear Array Model; 6.3 Performance Measures; 6.3.1 Noise Reduction; 6.3.2 Speech Distortion; 6.3.3 MSE Criterion; 6.4 Optimal Filters; 6.4.1 Maximum SNR; 6.4.2 Wiener; 6.4.3 MVDR; 6.4.4 Tradeoff; 6.4.5 LCMV; References; 7 A Bayesian Approach to the Speech Subspace Estimation; 7.1 Signal Model and Problem Formulation; 7.2 Estimation Based on the Minimum Mean-Square Distance; 7.3 A Closed-Form Solution Based on the Bingham Posterior 327 $aReferences8 Evaluation of the Time-Domain Speech Enhancement Filters; 8.1 Evaluation of Single-Channel Filters; 8.1.1 Rank-Deficient Speech Correlation Matrix; 8.1.2 Full-Rank Speech Correlation Matrix; 8.2 Evaluation of Multichannel Filters; References; Index 330 $a Speech enhancement is a classical problem in signal processing, yet still largely unsolved. Two of the conventional approaches for solving this problem are linear filtering, like the classical Wiener filter, and subspace methods. These approaches have traditionally been treated as different classes of methods and have been introduced in somewhat different contexts. Linear filtering methods originate in stochastic processes, while subspace methods have largely been based on developments in numerical linear algebra and matrix approximation theory. This book bridges the gap between the 606 $aSpeech processing systems 606 $aSignal processing 615 0$aSpeech processing systems. 615 0$aSignal processing. 676 $a006.454 700 $aBenesty$b Jacob$0721063 701 $aBenesty$b Jacob$0721063 801 0$bMiFhGG 801 1$bMiFhGG 906 $aBOOK 912 $a9910825222703321 996 $aSpeech enhancement$94076856 997 $aUNINA