Non-linear Adaptive Prediction of Speech with a Pipelined Recurrent Neural Network and a Linearised Recursive Least Squares Algorithm
نویسندگان
چکیده
A novel linearised Recursive Least Squares (LRLS) learning algorithm is presented for an adaptive non-linear forward predictor based on a Pipelined Recurrent Neural Network (PRNN). Simulation studies with speech signals show that the non-linear predictor does not perform satisfactorily when the previously proposed stochastic gradient (SG) algorithm is used. However, significantly improved results are demonstrated with the new LRLS algorithm. The non-linear structure affords prediction gains that are approximately 2dB higher than those of a linear structure RLS based predictor.
منابع مشابه
Nonlinear adaptive prediction of speech with a pipelined recurrent neural network
New learning algorithms for an adaptive nonlinear forward predictor that is based on a pipelined recurrent neural network (PRNN) are presented. A computationally efficient gradient descent (GD) learning algorithm, together with a novel extended recursive least squares (ERLS) learning algorithm, are proposed. Simulation studies based on three speech signals that have been made public and are ava...
متن کاملAdaptive Predictive Controllers Using a Growing and Pruning RBF Neural Network
An adaptive version of growing and pruning RBF neural network has been used to predict the system output and implement Linear Model-Based Predictive Controller (LMPC) and Non-linear Model-based Predictive Controller (NMPC) strategies. A radial-basis neural network with growing and pruning capabilities is introduced to carry out on-line model identification.An Unscented Kal...
متن کاملNon-linear Adaptive Prediction of Speech with a Pipelined Recurrent Neural Network and Advanced Learning Algorithms
متن کامل
Toward an optimal PRNN-based nonlinear predictor
We present an approach for selecting optimal parameters for the pipelined recurrent neural network (PRNN) in the paradigm of nonlinear and nonstationary signal prediction. Although there has recently been progress in terms of algorithms for training the PRNN, no account has been made of some inherent features of the PRNN. We therefore provide a study of the role of nesting, which is inherent to...
متن کاملDesign of an Intelligent Controller for Station Keeping, Attitude Control, and Path Tracking of a Quadrotor Using Recursive Neural Networks
During recent years there has been growing interest in unmanned aerial vehicles (UAVs). Moreover, the necessity to control and navigate these vehicles has attracted much attention from researchers in this field. This is mostly due to the fact that the interactions between turbulent airflows apply complex aerodynamic forces to the system. Since the dynamics of a quadrotor are non-linear and the ...
متن کامل