Abstract: An acoustic vibration sensor, also referred to as a speech sensing device, is provided. The acoustic vibration sensor receives speech signals of a human talker and, in response, generates electrical signals representative of human speech. The acoustic vibration sensor includes at least one diaphragm positioned adjacent to a front port and at least one coupler. The coupler couples a first set of signals to the diaphragm while isolating the diaphragm from the second set of signals. The coupler includes at least one material with acoustic impedance matched to the acoustic impedance of human skin.