HEAD MOUNTED MULTI-SENSORY AUDIO INPUT SYSTEM FOR REMOVING BACKGROUND SPEECH AND IMPROVING VARIABLE-RATE SPEECH CODING SYSTEMS AND FLOOR CONTROL IN REAL TIME COMMUNICATION

摘要

PURPOSE: A head mounted multi-sensory audio input system is provided to remove background speech and improve variable-rate speech coding systems and floor control in real time communication by generating a speech sensor signal based on an action undertaken by a speaker during speech, such as facial movement, bone vibration, throat vibration, throat impedance changes, etc. CONSTITUTION: An audio microphone(303) outputs a microphone signal based on a sensed audio input. A speech sensor(301) outputs a sensor signal based on a non-audio signal generated by speech movement. A speech detector component outputs a speech detection signal indicative of whether the user is speaking based on the sensor signal.