A new algorithm for parameter re-optimization in multi-pulse excitation LP synthesizer

Author

Chen, Yining ; Wang, Penghao ; Liu, Jia ; Liu, Runsheng

Author_Institution

Dept. of Electron. Eng., Tsinghua Univ., Beijing, China

fYear

2000

fDate

2000

Firstpage

560

Lastpage

563

Abstract

A new algorithm is proposed to re-optimize the synthesis filter parameters and pulse amplitudes in the Multi-Pulse Excitation Linear Prediction (LP) synthesizer. It is based on minimizing the perceptional weighted mean square errors between the original and the reconstructed speech. Steepest gradient descent and conjugate gradient descent are employed. The LP filter parameters and all the pulse amplitudes are iteratively re-optimized. Compared with the conventional multi-pulse excitation LPC algorithm, the new algorithm improves the segmental signal-to-noise ratio (SEGSNR) from about 20 dB to 23 dB. The informal listening also shows that the quality of synthesized speech is improved greatly

Keywords

conjugate gradient methods; filtering theory; gradient methods; mean square error methods; prediction theory; speech synthesis; 23 dB; LP filter parameters; conjugate gradient descent; linear prediction synthesizer; multi-pulse excitation LP synthesizer; parameter re-optimization; pulse amplitudes; reconstructed speech; segmental SNR; segmental signal/noise ratio; steepest gradient descent; synthesis filter parameters; synthesized speech; weighted mean square errors; Autocorrelation; Electronic mail; Equations; Iterative algorithms; Linear predictive coding; Mean square error methods; Nonlinear filters; Signal synthesis; Speech synthesis; Synthesizers;

fLanguage

English

Publisher

ieee

Conference_Titel

Circuits and Systems, 2000. IEEE APCCAS 2000. The 2000 IEEE Asia-Pacific Conference on

Conference_Location

Tianjin

Print_ISBN

0-7803-6253-5

Type

conf

DOI

10.1109/APCCAS.2000.913561

Filename

913561