A microphone array processing technique for speech enhancement in a reverberant space

摘要: In this paper, a new microphone array processing technique is proposed for blind dereverberation of speech signals affected by room acoustics. It based on the separate minimum-phase and all-pass components delay-steered multi-microphone signals. The are processed in cepstrum-domain, where spatial averaging followed low-time filtering applied. components, which contain source location information, frequency-domain performing retaining only component resulting output. underlying motivation processor to use spatio-temporal over single set synchronous segments from several microphones reconstruct speech, such that it applicable practical time-variant acoustic environments. Simulated impulse responses used evaluate compare conventional beamformer. Significant improvements gain important reductions reverberation listening tests observed.