There are a few threads about this subject, outlining the limitations and why it may not be possible.  Start here:

I am not sure giving users the ability to add all the speakers they want would be desirable. Good sound is about other things than stuffing rooms full of speakers.

It also presumes a processor/controller unit that either knows automatically, or can be programmed to define, what sounds to send to which speakers. The latter seems to me to be an option that’ll lead to lots of support calls!