The microphone array is the hidden gem of the Kinect sensor. The array is made up of four separate microphones spread out linearly at the bottom of the Kinect. By comparing when each microphone captures the same audio signal, the microphone array can be used to determine the direction from which the signal is coming. This technique can also be used to make the microphone array pay more attention to sound from one particular direction rather than another. Finally, algorithms can be applied to the audio streams captured from the microphone array in order to perform complex sound dampening effects to remove irrelevant background noise. All of this sophisticated interaction between Kinect hardware and Kinect SDK software allows speech commands to be used in a large room where the speaker’s lips are more than a few inches from the microphone.
KeywordsSpeech Recognition Automatic Gain Control Microphone Array Audio Stream Event Handler
Unable to display preview. Download preview PDF.