Sensory Increases Installs of Low-Power Sound Detector
Sensory's Low Power Sound Detector (LPSD) hardware for its TrulyHandsfree technology has been incorporated into chip designs from DSP Group, Knowles, and QuickLogic, with more chip partners to be announced.
LPSD allows chip manufactures whose products are suited for always-listening speech recognition system on consumer electronics devices to reduce the power requirements of their components, as well as the power demand of the entire speech recognition audio chain. Sensory's LPSD was designed to only wake the audio chain for a keyword trigger when potential speech is sensed. The solution conserves power usage by allowing the remaining audio chain of a device to be switched off until needed for speech processing.
"Sensory has been the industry's pioneer of low-power speech recognition technologies for more than 20 years, and while our line of speech recognition solutions have always been extremely power efficient, we created our Low Power Sound Detector technology for those looking to integrate always-listening voice trigger technology into ultra-low-power devices," said Todd Mozer, founder and CEO of Sensory, in a statement.
LPSD was developed as a supplemental solution designed to enable low-power devices to operate Sensory's TrulyHandsfree and other always-on, always-listening speech recognition technologies with minimal draw on the battery. A hardware implementation of LPSD reduces the overall system power for always-listening keyword trigger technology by 50 percent. Depending on the mix of speech to silence, Sensory's LPSD reduces processing requirements and lowers average power consumption.
"Sensory is the industry leader in voice processing systems for mobile applications," said Brian Faith, CEO of QuickLogic, in a statement. "Integration of Sensory's TrulyHandsfree and LPSD hardware with the QuickLogic EOS S3 multicore sensor processing SoC provides unprecedented always-on speech recognition capability and will facilitate a new generation of voice-driven applications capable of running on a fraction of the power consumption of competing technologies."
Sensory's voice recognition software powers a voice-processing microphone for waking up smart devices.
Partnership improves voice recognition, navigation, and control of the Vuzix M300 Smart Glasses.
TrulyHandsfree enables developers to add speech triggers and commands into apps with fewer than 20 lines of code.
Samsung's latest smartphone models offer voice trigger and speech recognition technology from Sensory.
TrulyHandsfree is now available on Intel low-power digital signal processors in Skylake, Anniedale, Broxton, Broadwell and Merrifield software-on-a-chip.