Title :
A DCT-based fast signal subspace technique for robust speech recognition
Author :
Huang, Jun ; Zhao, Yunxin
Author_Institution :
Beckman Inst. for Adv. Sci. & Technol., Illinois Univ., Urbana, IL, USA
fDate :
11/1/2000 12:00:00 AM
Abstract :
In this correspondence, a fast computational method is proposed to approximate the Karhunen-Loeve transform (KLT) for the covariance matrix of the autoregressive process. A fast algorithm which reduces the computation of eigenvalues of an N×N symmetric Toeplitz matrix from O(N3) in KLT to N2 is further developed. Experimental results demonstrate that the performance of the fast algorithm is very close to the KLT in eigenvalue computation and in energy constrained signal subspace speech enhancement for speech recognition in a car environment
Keywords :
Karhunen-Loeve transforms; Toeplitz matrices; autoregressive processes; covariance matrices; discrete cosine transforms; eigenvalues and eigenfunctions; speech enhancement; speech recognition; DCT-based fast signal subspace technique; KLT; Karhunen-Loeve transform; autoregressive process; car environment; covariance matrix; eigenvalues; energy constrained signal subspace speech enhancement; fast computational method; robust speech recognition; symmetric Toeplitz matrix; Covariance matrix; Discrete cosine transforms; Eigenvalues and eigenfunctions; Karhunen-Loeve transforms; Matrix decomposition; Robustness; Speech enhancement; Speech recognition; Subspace constraints; Symmetric matrices;
Journal_Title :
Speech and Audio Processing, IEEE Transactions on