Title :
A nonlinear adaptive predictor for speech compression
Author_Institution :
Dept. of Electr. & Comput. Eng., Puerto Rico Univ., Mayaguez, Puerto Rico
Abstract :
A neural nonlinear predictor for one dimensional signals is presented. It is based on a combination of linearization and QR decomposition that allows a fast adapting algorithm. The predictor is used in a speech compression algorithm that has proven to be superior to linear based models. The compression and training are done simultaneously, allowing the network to continually adapt to the signal. The results presented show that this algorithm outperforms a typical LPC coding algorithm
Keywords :
data compression; QR decomposition; linearization; neural nonlinear predictor; nonlinear adaptive predictor; one dimensional signals; speech compression; Bit rate; Compression algorithms; Joining processes; Linear predictive coding; Neural networks; Nonlinear equations; Predictive models; Speech coding; Stochastic resonance; Vector quantization;
Conference_Titel :
Neural Networks, 1996., IEEE International Conference on
Conference_Location :
Washington, DC
Print_ISBN :
0-7803-3210-5
DOI :
10.1109/ICNN.1996.549208
