We have identified some problems which might occur during the processing implementation:
->maintaining audio syncing
->video playback versus memory usage
->layering of video(tied with the playback depending on the method we use for it)
We have a solution for the audio. We will separate all audio from the videos and have the voice clips play continuously right from the start of the sketch. The only thing that need be controlled is the volume of each voice which depends on the number of videos showing of one particular face. This would mean a range of 0-6 (7 levels of volume). This could be done one of two ways, either controlling one sound clip at 7 different volume levels or muting/unmuting 6 repetitions of an identical voice clip.
For the other two issues we have not yet found a solution, or rather a method.