New high capacity optical discs and high bandwidth networks provide the capability for delivering multichannel audio.Although there are many one-and two-channel recordings in existence, only a handful of multichannel recordings exist.In this paper we propose a neural network approach that can synthesize microphone signals with the correct acousticalcharacteristics of specific venues that have been characterized in advance. These signals can be used to generate amultichannel recording with the acoustical characteristics of the original venue. The complex semi-cepstrum technique isemployed to extract features from musical signals recorded in a venue and these signals are sent into the fuzzy cerebellarmodel articulation controller (FCMAC) for training.
展开▼