AUDIO EQUIPMENT COMPRISING MEANS FOR DEBRISING A SPEECH SIGNAL BY FRACTIONAL TIME FILTERING, IN PARTICULAR FOR A HANDS-FREE TELEPHONY SYSTEM

摘要

The equipment has microphones (10, 12) for picking up a speech of a user of the equipment, and an adaptive filter (16), which is a fractional delay filter for modeling a delay shorter than a sampling period of a sampling unit. A voice activity detector (20) and a sensor (22) deliver signals representing presence or absence of the speech. The adaptive filter receives the signals representing presence or absence of speech so as to act selectively either to perform an adaptive search for filter parameters in the absence of speech or to freeze the parameters in the presence of speech. The adaptive filter is a filter having a least mean square (LMS) type linear prediction algorithm.