Syntiant Adds Smoke and Carbon Monoxide Alarm Detection to Glass-Break Sensor Model

Syntiant Corp., a leader in edge AI deployment, announced enhancements to the company’s smart home security solutions with the addition of real-time monitoring of smoke and carbon monoxide alarms to its popular glass-break audio deep learning-based model.

Beyond detecting the sound of breaking window glass, the new all-in-one smart home security solution can accurately listen for T3 (Temporal-3) audio alerts by smoke alarms and T4 (Temporal-4) audio alerts played by carbon monoxide detectors.

Recommended AI News: Riding on the Generative AI Hype, CDP Needs a New Definition in 2024

“Now within a single device, this new bundled solution keeps homes safe and secure without having to replace an entire security alert system,” said Kurt Busch, CEO of Syntiant. “Notifications are sent in real time and this all-in-one solution maintains the same industry-leading accuracy Syntiant customers rely upon, while reducing power draw by a factor of five, dramatically increasing battery life, privacy and responsiveness. More features, at lower power, means a better smart home end-user experience with longer-lasting devices.”

Recommended AI News: Acer Unveils New Swift Go AI PCs With Intel Core Ultra Processors

Syntiant used real-world data alongside a wide variety of synthetic background noises to train the machine learning models to discern between sounds like the breaking of a window to glass shattering in a sink. Environments are passively monitored to reduce false wakeups from other sounds, noises and anomalies, extending battery life in the process.

Ultra-Low-Power Hardware

Deployed using the Syntiant NDP120 Neural Decision Processor™, the new all-in-one glass-break/T3/T4 model can run for five years on three AA batteries while detecting glass breaking, smoke alarms and carbon monoxide alarms. Packaged with the Syntiant Core 2™, the company’s 2nd generation, highly flexible, ultra-low-power DNN, the NDP120 supports Ok Google and Hey Google hotwords at under 280uW and can run multiple applications simultaneously at under 1mW.

Recommended AI News: AONDevices Boosts Edge AI Global Presence With Insightful Partnership

[To share your insights with us, please write to sghosh@martechseries.com]