How to Optimize the Gain Filter of LD-CELP

Author

Gang, Zhang ; Xie Kerning ; Zhefeng, Zhao ; Chunyu, Xue

Author_Institution

Coll. of Inf. Eng., Taiyuan Univ. of Technol., Shanxi Taiyuan

Volume

1

fYear

0

fDate

0-0 0

Firstpage

1206

Lastpage

1211

Abstract

The recommendation G.728 depends on the Levinson-Durbin algorithm to update gain filter coefficients. In this paper, it is introduced by three different methods which are the weighted L-S recursive filter, the finite memory recursive filter and the BP neural network, respectively. Because quantizer has not existed at optimizing gain filter, the quantization SNR can not be used to evaluate its performance. This paper proposes a scheme to estimate SNR so that the gain predictor can be separately optimized with the quantizer. Using these three gain filter the speech coding results are all better than the G.728. The weighted L-S algorithm has the best effect. Its average segment SNR is higher than the G.728 about 0.76dB. It is also used to evaluate the case that excitation vector is 16 and 20 samples respectively; the weighted L-S algorithm has similarly the best result

Keywords

backpropagation; neural nets; quantisation (signal); recursive filters; speech coding; G.728 recommendation; Levinson-Durbin algorithm; SNR estimation; backpropagation neural network; excitation vector; finite memory recursive filter; gain filter coefficient; gain filter optimization; gain predictor; low delay code excited linear prediction; quantizer; signal to noise ratio; speech coding; weighted least square; Adaptive filters; Educational institutions; Neural networks; Optimization methods; Performance evaluation; Performance gain; Quantization; Signal processing; Speech coding; Testing;

fLanguage

English

Publisher

ieee

Conference_Titel

Information and Communication Technologies, 2006. ICTTA '06. 2nd

Conference_Location

Damascus

Print_ISBN

0-7803-9521-2

Type

conf

DOI

10.1109/ICTTA.2006.1684548

Filename

1684548