Forecasting Sound Impulse Response Using Neural Network
To improve the computation of sound impulse response for the virtual audio environment technique, a neural network approach has been developed to forecast the sound impulse response from the computed data obtained with simulation software. A three-layered neural network was used. Methodology of the approach and optimum architecture of the network are described. Application examples are presented and compared with practical response.
This paper costs $20 for non-members, $5 for AES members and is free for E-Library subscribers.