A voice acquisition system for a vehicle includes an interior rearview mirror assembly attached at an inner portion of the vehicle. The mirror assembly preferably includes at least one microphone for receiving audio signals within a cabin of the vehicle and generating an output signal indicative of these audio signals, the at least one microphone providing sound capture for at least one of a hands free cell phone system, an audio recording system and an emergency communication system. The system includes a control that receives the output signal from the at least one microphone and distinguishes the presence of vocal signals from non-vocal signals. The system preferably comprises a learning mode whereby a vocal characteristic of a particular driver or occupant of the vehicle is learnt so that the ratio of vocal signals emanating from that particular driver or occupant to non-vocal noise signals received by the system is enhanced.