hi all,
Can you please tell me some algorithms which can be implemented to exploit the correlation between the channels of a multi-channel audio and encode it more compactly.
I know that the simple things are mid-diff, mid-side and side-diff codings of stereo. But my interests are in 5.1 channels to 10.2 channels. One way is to code the mid-diff and other codings i mentioned above for Front Right and Front Left, Back Right and Back Left, Aux. Right and Aux Left and so on. But Do u have more precise algorithms to be used to do multi-channel coding.
Please reply on this topic....

Regars
DLS

The more advanced lossless coders all use adaptive predictors to do their filtering. Simply using samples from all channels in the input windows for these predictors is the best method.