Voice recognition by sensor fusion; radar plus MEMS microphones
Together, the two sensing techniques guarantee optimal sound recognition and flawless execution for digital voice assistance across a broad range of voice-controlled devices. Target applications are “especially smart” home, as well as smart TV and set top box, secure keyless entry systems, and other voice-operated consumer devices.
The performance of current MEMS microphone limits the effectiveness of systems: when multiple persons are speaking, the veracity and location of the voice source cannot be precisely identified and separated from inanimate object noise. Infineon’s 60 GHz 2Tx/4Rx radar IC with accompanying antenna and the 70 dB SNR microphone helps to overcome these impediments. The microphones are based on Infineon’s dual backplate MEMS technology, suited to far-field voice capture and beamforming. Improving the SNR of the microphone will further enhance the performance.
The XMOS audio processor analyses the signal data from an array of Infineon’s digital microphones, adjusting the angular and distance data from each microphone to produce a beam [of sensititvity] at an angle identified by the radar data. A lit LED indicates presence detection and where sound is taken from.
The combination of an Infineon radar and the XMOS beamformer allows the microphone to target a specific object precisely, even with the object moving and indistinctive noise. The microphone delivers a superior user experience compared to existing voice-controlled systems, with clear understanding of commands and flawless execution that prevents unwanted activations of devices.
Infineon; www.infineon.com/audiobeamforming
XMOS; www.xmos.com
If you enjoyed this article, you will like the following ones: don't miss them by subscribing to :
