Hi I did mfcc of multiparticipant voices recording. now I have the mfcc representations of me and 2 another friends.
now as i said, i need to find with the optimization problem what are the minimal weigths when we subtract the friends recording vectors from the group multiparticipant conversation vectors.
the weights are the results that i need. I thought about dwt or something like this but i don't know what is the exactly method.