Beam forming for a wearable computer

A wearable computer comprising:

• one or more processors;
• one or more memories storing instructions executable by the one or more processors;
• a microphone array comprising a plurality of microphones coupled to the one or more processors and configured to generate a plurality of audio signals from an utterance detected by the microphone array, wherein the utterance is from a user associated with the wearable computer;
• one or more motion sensors coupled to the one or more processors and configured to generate motion data in response to movement of the wearable computer;
• a localization module stored in the one or more memories and executable by the one or more processors to:
• determine an activity among one or more predetermined activities based at least in part on the motion data;
• and
• determine, based at least in part on the activity, a direction vector pointing in a direction from the wearable computer towards a mouth of the user;
• and
• a beamforming module stored in the one or more memories and executable by the one or more processors to determine parameters corresponding to a beampattern having a direction in substantial alignment with the direction vector, wherein the parameters are to be used to obtain a beamformed audio signal.

A method comprising:

• generating, using a microphone array of a wearable computer, a plurality of audio signals from an utterance detected by the microphone array, wherein the utterance is from a user associated with the wearable computer;
• generating, by one or more motion sensors of the wearable computer, motion data based on movement of the wearable computer;
• determining, based at least in part on the motion data, a direction of a mouth of the user relative to the wearable computer;
• determining a confidence level of the direction of the mouth of the user;
• determining parameters corresponding to a beampattern having:
• a direction in substantial alignment with the direction of the mouth of the user to focus the microphone array toward the mouth of the user;
• and
• a width that is based at least in part on the confidence level;
• and
• obtaining a beamformed audio signal using the beampattern.

A method comprising:

• generating, using a microphone array of a wearable computer, a plurality of audio signals from an utterance detected by the microphone array, wherein the utterance is from a user associated with the wearable computer;
• generating, by one or more motion sensors of the wearable computer, motion data based on movement of the wearable computer;
• determining, based at least in part on the motion data, an activity among one or more predetermined activities;
• determining, based at least in part on the activity, a direction of a mouth of the user relative to the wearable computer;
• determining parameters corresponding to a beampattern having a direction in substantial alignment with the direction of the mouth of the user to focus the microphone array toward the mouth of the user;
• and
• obtaining a beamformed audio signal using the beampattern.