Approximated Correlation Matrix and Pulse Prediction for Fast Algebraic Code-Excited LPC Speech Coders

Author

Tsai, Shu-Min ; Yang, Jar-Ferr

Author_Institution

Dept. of Electr. Eng., Nat. Cheng Kung Univ., Tainan

fYear

2006

fDate

9-12 July 2006

Firstpage

181

Lastpage

184

Abstract

With reducing computational complexity, an approximated correlation matrix of the vocal impulse response is proposed in algebraic-code-excited linear prediction (ACELP) coders. By exploring statistical characteristics, we only need to calculate a small portion of correlation coefficients before ACELP search procedure. If we further combine a pulse position prediction algorithm, we can reduce the arithmetic complexity in pre-computing autocorrelation matrix and the number of pulse position combinations with imperceptible degradation in speech quality performance. The proposed scheme can be applied to all ACELP coders such as ITU G.729 and G.723.1

Keywords

algebraic codes; approximation theory; correlation methods; linear predictive coding; matrix algebra; search problems; speech coding; statistical analysis; transient response; ACELP search procedure; algebraic-code-excited linear prediction coder; approximated correlation matrix; pulse position prediction algorithm; speech quality performance; statistical characteristics; vocal impulse response; Arithmetic; Autocorrelation; Bit rate; Computational complexity; Linear predictive coding; Matrices; Mean square error methods; Prediction algorithms; Search methods; Speech synthesis;

fLanguage

English

Publisher

ieee

Conference_Titel

Multimedia and Expo, 2006 IEEE International Conference on

Conference_Location

Toronto, Ont.

Print_ISBN

1-4244-0366-7

Electronic_ISBN

1-4244-0367-7

Type

conf

DOI

10.1109/ICME.2006.262599

Filename

4036566