Wake up and listen: Vesper quiescent-sensing MEMS device innovation

July 04, 2016 // By Graham Prophet
I have a strong belief that the most natural and efficient way to communicate with devices, in the coming of the Internet of Things (IoT), is the human voice. The primary element for this effort to be successful is the microphone and the primary features needed in such a system are low power, small size, rugged construction, and excellent signal-to-noise capability.

Vesper is a privately held piezoelectric MEMS company which began at the University of Michigan; it has improved the acoustic experience with a wide range of consumer products via a patented piezoelectric MEMS platform. These products include smartphones, smart speakers, Internet of Things (IoT) devices and connected automobiles.

Vesper has now demonstrated the first commercially available quiescent-sensing MEMS device, providing designers the possibility of acoustic event-detection devices at virtually zero power draw at 3 µA of current while in listening mode. This piezoelectric MEMS microphone — VM1010 — will allow designers to advance voice and acoustic event monitoring in their systems. See Figure 1 for the pin out of this IC.

 

Figure 1. The pin out of the VM1010 (Image courtesy of Vesper)

 

Matt Crowley, Vesper CEO told me that this quiescent-sensing MEMS microphone is the only device that uses sound energy itself to wake a system from full power-down. It is known that even when fully powered-off, batteries in smartphones and smart speakers naturally dissipate 40-80 µA, which is far more current than this device needs. So a design using this technology will see no difference in battery-life for a system using the VM1010 in listening mode and a fully powered-down conventional system. See Figure 2 for specifications.

 

Figure 2. Key specifications for the VM1010 (Image courtesy of Vesper)

 

Even in sleep mode, this microphone preserves its very high signal-to-noise ratio (SNR) — which is needed for a wider acoustic range. In fact, there is absolutely no SNR penalty from going into low power, extending the distance of far field voice control in battery-powered systems. Watch the following YouTube video ( here) which shows the actual performance and capabilities like wake-on-sound.

 

When a designer embeds this device into a voice-powered TV remote control or smart speaker, it could allow you to turn on your device from across the room without having to