Rate-distortion modeling of spherical vector quantization performance in transform excitation audio coding

Author

Patchoo, Wisarn ; Fischer, Thomas R.

Author_Institution

Sch. of EECS, Washington State Univ., Pullman, WA, USA

fYear

2010

fDate

19-21 May 2010

Firstpage

1075

Lastpage

1079

Abstract

A block-based Gaussian mixture model (GMM) is used to model the distribution of the transform audio data encoded using spherical vector quantization and lossless coding. The expectation-maximization algorithm is used to design the GMM to model the marginal density of the transform coefficients and the block energy density. A GMM-based rate-distortion function is derived and shown to closely match the observed spherical VQ performance.

Keywords

Gaussian processes; audio coding; expectation-maximisation algorithm; rate distortion theory; vector quantisation; block energy density; block-based Gaussian mixture model; expectation-maximization algorithm; lossless coding; marginal density; rate-distortion function; rate-distortion modeling; spherical vector quantization performance; transform audio data encoded; transform coefficients; transform excitation audio coding; Algorithm design and analysis; Audio coding; Discrete Fourier transforms; Expectation-maximization algorithms; Laplace equations; Lattices; Performance loss; Probability density function; Rate-distortion; Vector quantization; Gaussian distribution; audio coding; rate distortion theory; vector quantization;

fLanguage

English

Publisher

ieee

Conference_Titel

Electrical Engineering/Electronics Computer Telecommunications and Information Technology (ECTI-CON), 2010 International Conference on

Conference_Location

Chaing Mai

Print_ISBN

978-1-4244-5606-2

Electronic_ISBN

978-1-4244-5607-9

Type

conf

Filename

5491641