<--- Back to Details
First PageDocument Content
Statistical classification / Linear discriminant analysis / Singular value decomposition / Projection / Speech recognition / Hidden Markov model / Eigenvalues and eigenvectors / Degrees of freedom / Statistics / Algebra / Multivariate statistics
Date: 2006-12-20 19:07:22
Statistical classification
Linear discriminant analysis
Singular value decomposition
Projection
Speech recognition
Hidden Markov model
Eigenvalues and eigenvectors
Degrees of freedom
Statistics
Algebra
Multivariate statistics

LINEAR FEATURE SPACE PROJECTIONS FOR SPEAKER ADAPTATION George Saon, Geoffrey Zweig and Mukund Padmanabhan IBM T. J. Watson Research Center, Yorktown Heights, NY, 10598 E-mail: fgsaon,gzweig,, Phone: (9

Add to Reading List

Source URL: research.microsoft.com

Download Document from Source Website

File Size: 64,30 KB

Share Document on Facebook

Similar Documents

New low-cost microsystem-based robot force sensor / Kostengünstiger Kraftsensor für Roboter auf Mikrosystem-Basis English SMErobot™ presents a new six Degrees Of Freedom (DOF) force/torque sensor technology developed

New low-cost microsystem-based robot force sensor / Kostengünstiger Kraftsensor für Roboter auf Mikrosystem-Basis English SMErobot™ presents a new six Degrees Of Freedom (DOF) force/torque sensor technology developed

DocID: 1vrOL - View Document

Betatron Motion with Coupling of Horizontal and Vertical Degrees of Freedom – Part II S.A. Bogacz, G.A. Krafft, S. DeSilva and R. Gamage Jefferson Lab and Old Dominion University

Betatron Motion with Coupling of Horizontal and Vertical Degrees of Freedom – Part II S.A. Bogacz, G.A. Krafft, S. DeSilva and R. Gamage Jefferson Lab and Old Dominion University

DocID: 1uypj - View Document

1402  PROCEEDINGS OF THE IEEE, VOL. 65, NO. 9 , SEPTEMBER 1977 the recursive fdter d o e s not have enpugh degrees of freedom to represent the Wiener filter) then d and b will not, in general, converge to a and b since t

1402 PROCEEDINGS OF THE IEEE, VOL. 65, NO. 9 , SEPTEMBER 1977 the recursive fdter d o e s not have enpugh degrees of freedom to represent the Wiener filter) then d and b will not, in general, converge to a and b since t

DocID: 1un7i - View Document

ISAMMDoF2015 International Symposium on Advanced Materials Having Multi-Degrees-of-Freedom – Optical properties, Structural Analyses, Imaging and Informatics of Materials –

ISAMMDoF2015 International Symposium on Advanced Materials Having Multi-Degrees-of-Freedom – Optical properties, Structural Analyses, Imaging and Informatics of Materials –

DocID: 1u37B - View Document

Betatron Motion with Coupling of Horizontal and Vertical Degrees of Freedom – Part I S.A. Bogacz, G.A. Krafft, S. DeSilva and R. Gamage Jefferson Lab and Old Dominion University

Betatron Motion with Coupling of Horizontal and Vertical Degrees of Freedom – Part I S.A. Bogacz, G.A. Krafft, S. DeSilva and R. Gamage Jefferson Lab and Old Dominion University

DocID: 1tRWu - View Document