Multichannel Audio Modeling and Coding Using a Multiband Source/Filter Model

In this paper we propose a source/filter model for achieving low bitrate transmission of multichannel audio signals, in which the filter part corresponds to the specifics of each microphone information while the source part contains mostly the interchannel similarities. Using the appropriate filter for each channel and the source part of only one of the microphone signals, we can resynthesize a high quality approximation of each channel; thus, the filter part of each channel need only be encoded. Low datarates can be achieved in the order of few KBits/sec/channel focusing on applications such as remote mixing or distributed musicians collaboration.