Title :
Nonlinear Long-Term Prediction of Speech Based on Truncated Volterra Series
Author :
Despotovic, Vladimir ; Goertz, Norbert ; Peric, Zoran
Author_Institution :
Tech. Fac. of Bor, Univ. of Belgrade, Bor, Serbia
fDate :
3/1/2012 12:00:00 AM
Abstract :
Previous studies of nonlinear prediction of speech have been mostly focused on short-term prediction. This paper presents long-term nonlinear prediction based on second-order Volterra filters. It will be shown that the presented predictor can outperform conventional linear prediction techniques in terms of prediction gain and “whiter” residuals.
Keywords :
Volterra series; nonlinear filters; speech processing; conventional linear prediction technique; nonlinear long-term prediction; prediction gain residual; second-order Volterra filter; short-term prediction; truncated Volterra series based speech; whiter residual; Computational modeling; Correlation; Kernel; Maximum likelihood detection; Nonlinear filters; Predictive models; Speech; Linear predictive coding; long-term prediction; nonlinear filters; volterra series;
Journal_Title :
Audio, Speech, and Language Processing, IEEE Transactions on
DOI :
10.1109/TASL.2011.2169788
