As per patent application by the US Patent and Trademark Office, Apple has been investigating to enhance headphone device which can recognize the user’s voice along with multiple on board microphones, turning on noise suppression systems besides directing beam forming mics towards a user’s head. These new generation headphones will come with microphones to automatically record voice output when the user starts talking which will also come with noise cancellation feature by default to subdue the background noise due to the second microphone on the headphones.
Headphone featuring built-in Accelerometer
It will also have microphones on the wires just as a standard handset and the earphones can automatically judge when the user begins talking due to the beamforming microphones. Besides this, the headphone featuring the built-in accelerometer will also judge when the user begins talking with the vibration created through the vocal chords which means that the microphone will be active to pick up voice from the user at any point of time.
Moreover the user will also get access to the device with the help of the Siri features which could be very helpful while driving or performing other activities. Presently most of the users depend on headphones for voice calls as well as music and Apple has brought about a change with its enhanced pair of headphones.
Sensor, Voice Activity Detector
Its system of working begins with the accelerometer and in some embodiment; the sensor is utilized as a voice activity detector – VAD to detect vibration generated by the user. The voice speech both generated by the user’s vocal chords as well as the unvoiced speech generated without the vocal chords are detected by the accelerometer through vibration of the tissue and bone.
The detected accelerometer vibration data is then joined by signals from the front and rare facing microphones which are embedded on one or both of the ear buds. The accelerometer in some embodiments is tuned to detect lower frequency vibration known as VADa system and alternatively, from the two microphones signals output could be used in determining voiced speech or a VADm arrangement.
Effective Beamforming Algorithm
The VAD system functions by scanning both voiced –VADv as well as unvoiced –VADu signals which are combined with the other signals for noise suppression circuit. Depending on the information gathered, noise from the area around the user could be estimated with an effective beamforming algorithm which can be applied to the system’s various microphone arrays.
Mics located in the headphones in some embodiments and along the unit’s wire were set to a general beamform focused at the user’s face. Besides this a more active beamformer is described which can be adaptively steered based on VAD output while alternatively a noise suppression system could also be made effective by steering microphones towards the ambient noise which may not be associated with voiced or unvoiced speech that is removed from the signal.
