A SIGNAL SUBSPACE TRACKING ALGORITHM FOR MICROPHONE ARRAY-PROCESSING OF SPEECH

This paper presents a method of adaptive microphone array beamforming using matched filters with signal subspace tracking, Our objective is to enhance near-field speech signals by reducing multipath and reverbe ration, In real applications such as speech acquisition in acoustic en vironments, sources do not propagate along known and direct paths. Par ticularly in hands-free telephony, we have to deal with undesired prop agation phenomena such as reflections and reverberation, Prior methods developed adaptive microphone arrays for noise reduction after a time delay compensation of the direct path, This simple synchronization is insufficient to produce an acceptable speech quality, and makes adapt ive beamforming unsuitable, In this contribution, we prove the identif ication of source-to-array impulse responses to be possible by subspac e tracking, We consequently show the advantage of treating synchroniza tion as a matched filtering step, Speech quality is indeed enhanced at the output by the suppression of reflections and reverberation (i.e., dereverberation), and efficient adaptive beamforming for noise reduct ion is applied without risk of signal cancellation, Evaluations confir m the performance achieved by the proposed algorithm under real condit ions.