Research On Speech Enhancement Method Based On Spherical Microphone Array

In actual environment, speech is inevitably affected by the transmission noise, internalnoise and other interference, so that the signal received by the microphone is noise speechsignal. Thus, not only the speech quality is affected, but also the performance of speechprocessing system is become worse. Therefore, it’s needed to introduce the speechenhancement technique to remove the noise, and to restore the original speech.Microphone array noised speech processing method is one of the most commonlymethod of speech enhancement and the spatial information of the signal can be effective tosuppress the noise and interference, with the good performance achieved. One or twodimensional array configurations are usually used in the microphone array. Considering theadvantages of spherical microphone array such as symmetry, rotating characteristics andspherical harmonic orthogonality, a research on speech enhancement based on sphericalmicrophone array is developed in this paper.The characteristics of the microphone array and the beam forming method are analyzedin the paper. Method for the spherical microphone array which is based on the traditionalmicrophone array is introduced. Considering the acoustic environment when the desiredsource and interference have the same direction and frequency, spatial radial canceller filterwhich combined the characteristics of dolph-chebyshev radial filter and radial notch filter ispresented in the paper. The simulation results show that not only the characteristics ofdolph-chebyshev radial filter is well inherited by the spatial radial canceller filter, but also thespatial radial canceller filter has better performance on speech enhancement.