Techday
This page was machine-translated and may differ from the original. View original

What voice processing semiconductor technology is used in speech-understanding microphones?

Google 우선 소스Published2017.03.15 12:49
ST Unveils Detailed Technology Providing Keyword Recognition Capabilities in Small Package
DSP Group collaborates on ultra-low power voice processing chips and sensors

What semiconductor technology is hidden within voice-operated smart speakers, TV remotes, and smart home systems?

STMicroelectronics, together with its DSP group and Sensory, has unveiled details on a power-efficient voice detection and voice processing microphone that provides keyword recognition capabilities in a small package. The small package device integrates a low-power ST MEMS microphone based on the DSP Group's ultra-low-power voice processing chip and Sensory's voice recognition firmware, and the packaging utilizes ST technology.




Typically, wake-on-sound microphones have limited processing power and require the main system processor to recognize commands; however, ST's microphones utilize the computational capabilities of the DSP group to recognize commands without activating the main system. STMicrophone solutions employ the DSP group's HD Clear ultra-low power audio processing chip to significantly reduce energy consumption, extending equipment lifespan by years without the need for recharging or replacement.

"It is not simply about listening to voice; it can instantly understand commands without utilizing the main processor's power and computational resources," said an ST Sensors representative. "This smart integration is key to implementing voice interfaces added to IoT products and applications, including Industry 4.0."
본 기사에 대한 정정·반론·추후보도 청구는 보도 청구 안내를, 그간 게재된 보도문은 정정·반론보도 모아보기를 참고해 주세요.
김자영 기자