Sensory launches always-listening edge AI for smart home devices
Sensory has launched a new software product to embed always-listening AI into consumer electronics, which can identify common and critical sounds within the home for enhanced security without the privacy risk of constantly sending audio recordings to the cloud.
TrulySecure Sound ID is capable of recognizing glass breaking, babies crying, dogs barking, home security alarms, smoke and CO detector alarms and low-battery warnings, and more, and enables devices to send alerts and sound clips to residents to help them understand the situation. All analysis is performed on-device with Sensory’s ‘AI at the edge’ technology.
“With so many voice-controlled products entering our homes, we saw an excellent opportunity to enable the microphones in these devices to do more than just listen for wake words and recognize speech,” said Todd Mozer, CEO of Sensory. “With TrulySecure Sound ID, we’re making it possible for companies to create smart always-listening products that can alert us of things happening within our homes that we may not hear, and safeguard our home and family from potentially dangerous situations.”
Sound ID is offered as part of Sensory’s TrulySecure Speaker Verification (TSSV) suite for smart speakers and IoT devices, which also provides both text dependent and text independent voice biometrics.
Sensory upgraded TSSV with text independent speaker identification and accuracy improvements last year, and also added support for 3D cameras and cloud-based biometrics to its TrulySecure 4.0 SDK in October.