Title :
The application of BP type neural network in speech coding
Author_Institution :
Taiyuan Univ. of Technol., China
Abstract :
The paper discusses the method that uses the BP algorithm of neural networks to code the speech signal by means of nonlinear prediction aiming at the non-stationary and nonlinear features of speech signals. Taking the G.721 ADPCM speech coding algorithm as an example, the method of nonlinear prediction replacing linear prediction is presented. The results indicate that the coding quality using neural network nonlinear prediction is better than that of using linear prediction
Keywords :
backpropagation; differential pulse code modulation; neural nets; prediction theory; speech coding; G.721 ADPCM speech coding; backpropagation; neural network; nonlinear prediction coding; speech coding; Intelligent networks; Neural networks; Speech coding;
Conference_Titel :
Intelligent Control and Automation, 2000. Proceedings of the 3rd World Congress on
Conference_Location :
Hefei
Print_ISBN :
0-7803-5995-X
DOI :
10.1109/WCICA.2000.862486
