Nonlinear perceptual audio filtering using support vector machines

Author

Hill, Simon I. ; Wolfe, Patrick J. ; Rayner, Peter J W

Author_Institution

Dept. of Eng., Cambridge Univ., UK

fYear

2001

fDate

6/23/1905 12:00:00 AM

Firstpage

488

Lastpage

491

Abstract

The perceptually based loss functions for audio filtering used by P.J. Wolfe and S.J. Godsill (see Proc. IEEE ICASSP, vol.2, p.821-4, 2000) are shown to fit well within a complex-valued support vector machine (SVM) framework. SVM regression is extended to the estimation of complex-valued functions, including the derivation of a variant of the sequential minimal optimisation (SMO) algorithm. Audio filters are derived using this, based on an autoregressive (AR) model used for audio and two different Hermitian kernel functions. Results are found to be promising, and further improvements are discussed

Keywords

audio signal processing; autoregressive processes; estimation theory; filtering theory; learning automata; nonlinear filters; optimisation; statistical analysis; Hermitian kernel functions; autoregressive model; complex-valued function estimation; loss functions; nonlinear perceptual audio filtering; sequential minimal optimisation algorithm; support vector machines; Discrete Fourier transforms; Filtering; Filters; Frequency; Kernel; Probability density function; Signal processing; Signal processing algorithms; Speech; Support vector machines;

fLanguage

English

Publisher

ieee

Conference_Titel

Statistical Signal Processing, 2001. Proceedings of the 11th IEEE Signal Processing Workshop on

Print_ISBN

0-7803-7011-2

Type

conf

DOI

10.1109/SSP.2001.955329

Filename

955329