Sound Source Seperation with Distributed Microphone Arrays in the Presence of Clocks Synchronization Errors

We first present our studies on how clock synchronization errors affect the performance of sound source separation with a distributed microphone array. We show that our previously proposed energy-based sound source separation method is robust to constant clock shift errors but is more sensitive to clock drift errors. We then propose a novel technique to address the clock drift errors. The key observation is that as the amount of clock drift increases, so does the correlation between the energies of the separated sources which are obtained from the Independent Component Analysis (ICA). Based on this observation, we propose an optimization technique to solve the clock drifting parameter. Experiment results are shown to validate our approach.