Blind source separation based on acoustic pressure distribution and normalized relative phase using dodecahedral microphone array

We developed a small dodecahedral microphone array device and propose a sound source separation method based on frequency-domain independent component analysis with the developed device. The developed device´s diameter is 8 cm and the intervals among each face are 36°. Microphones can be installed on ten faces except the top and bottom faces, and 16 holes exist on each face. Our proposed method solves the permutation problem, which is frequency-domain independent component analysis´s difficult problem, with acoustic pressure distribution that was observed in the device´s faces and the normalized relative phases at each microphone in the high and low frequency ranges, respectively. In our experiments, three kinds of mixture signals were used. The separation performances were evaluated by the signal-to-interference rate improvement score and compared with the conventional method and the ideal condition. The results indicate that the proposed method using the developed device is effective.