This letter addresses a neural network (NN)-based predictor for the LP (Linear Prediction) residual. A new NN predictor takes into consideration not only prediction error but also quantization effects. To increase robustness against the quantization noise of the nonlinear prediction residual, a constrained back propagation learning algorithm, which satisfies a Kuhn-Tucker inequality condition is proposed. Preliminary results indicate that the prediction gain of the proposed NN predictor was not seriously decreased even when the constrained optimization algorithm was employed.
The copyright of the original papers published on this site belongs to IEICE. Unauthorized use of the original or translated papers is prohibited. See IEICE Provisions on Copyright for details.
Copy
Ki-Seung LEE, "Nonlinear Long-Term Prediction of Speech Signal" in IEICE TRANSACTIONS on Information,
vol. E85-D, no. 8, pp. 1346-1348, August 2002, doi: .
Abstract: This letter addresses a neural network (NN)-based predictor for the LP (Linear Prediction) residual. A new NN predictor takes into consideration not only prediction error but also quantization effects. To increase robustness against the quantization noise of the nonlinear prediction residual, a constrained back propagation learning algorithm, which satisfies a Kuhn-Tucker inequality condition is proposed. Preliminary results indicate that the prediction gain of the proposed NN predictor was not seriously decreased even when the constrained optimization algorithm was employed.
URL: https://global.ieice.org/en_transactions/information/10.1587/e85-d_8_1346/_p
Copy
@ARTICLE{e85-d_8_1346,
author={Ki-Seung LEE, },
journal={IEICE TRANSACTIONS on Information},
title={Nonlinear Long-Term Prediction of Speech Signal},
year={2002},
volume={E85-D},
number={8},
pages={1346-1348},
abstract={This letter addresses a neural network (NN)-based predictor for the LP (Linear Prediction) residual. A new NN predictor takes into consideration not only prediction error but also quantization effects. To increase robustness against the quantization noise of the nonlinear prediction residual, a constrained back propagation learning algorithm, which satisfies a Kuhn-Tucker inequality condition is proposed. Preliminary results indicate that the prediction gain of the proposed NN predictor was not seriously decreased even when the constrained optimization algorithm was employed.},
keywords={},
doi={},
ISSN={},
month={August},}
Copy
TY - JOUR
TI - Nonlinear Long-Term Prediction of Speech Signal
T2 - IEICE TRANSACTIONS on Information
SP - 1346
EP - 1348
AU - Ki-Seung LEE
PY - 2002
DO -
JO - IEICE TRANSACTIONS on Information
SN -
VL - E85-D
IS - 8
JA - IEICE TRANSACTIONS on Information
Y1 - August 2002
AB - This letter addresses a neural network (NN)-based predictor for the LP (Linear Prediction) residual. A new NN predictor takes into consideration not only prediction error but also quantization effects. To increase robustness against the quantization noise of the nonlinear prediction residual, a constrained back propagation learning algorithm, which satisfies a Kuhn-Tucker inequality condition is proposed. Preliminary results indicate that the prediction gain of the proposed NN predictor was not seriously decreased even when the constrained optimization algorithm was employed.
ER -