According to the company, the new VM2000 piezoelectric MEMS microphone is "the only MEMS microphone that maintains high system performance over the life of a product." Targeting voice-activated systems - such as smart speakers, smart homes, and Internet of Things (IoT) products — that use microphone arrays to enhance voice capture and clarity, the VM2000 is offered as solving a critical need for stable, reliable, and accurate microphone arrays that do not degrade over time.
"The growth of voice-enabled smart speakers — which use up to seven microphones per device — as well as smart home automation systems and diverse IoT products is spurring demand for MEMS microphone arrays," says Matt Crowley, CEO, Vesper. "The problem with capacitive MEMS microphones, however, is that they degrade over time when exposed to contaminants. If one microphone in the array drifts or gets wet, dirty or dusty, the entire microphone array will fail, severely limiting acoustic system performance and far-field accuracy."
"Only piezoelectric MEMS microphones such as the VM2000 remain stable and reliable throughout a product's lifetime. This ensures long-term acoustic performance of voice-activated systems built with Vesper microphones."
Offered as the company's highest-performance MEMS microphone to date, the VM2000 features a 66-dB signal-to-noise ratio (SNR) and an acoustic overload point (AOP) of 133-dB sound pressure level (SPL) - supporting clear speech recognition in loud environments. In addition, says the company, its piezoelectric microphones cut material costs by up to 50 percent, enabling them to cost much less than capacitive microphones.
The VM2000 measures 3.76 x 2.95 x 1.3 mm and is reflow solder-compatible with no sensitivity degradation. It is currently sampling and will be in volume production in Q1 2018.
MEMS mic pioneer gets investment from Amazon, China
Smart speakers driving adoption of smart home devices
Audio control energizing microphone, speaker markets
Home audio market: 'It's all about the voice'
Choosing a far-field voice recognition DSP