Sensory Unveils Next-Generation SoundID: A Breakthrough in Low-Power, High-Precision Sound Recognition

Sensory, leaders in sound detection technology, today announced a comprehensive update to their pioneering product, SoundID. The two-pronged innovation focuses on enhancing the system's low-power sound detection capabilities, along with a significant expansion of its cloud-based sound detection model.

The newly enhanced low-power SoundID uses state-of-the-art machine learning technology to detect sound events while optimizing power consumption.The compact models maintain a size of around 100kBytes, ensuring high accuracy with minimum power consumption. SoundID can operate on high-performance, low-power DSP chips based on the Cadence HIFI5 core, permitting real-time analysis and accurate identification of sound events like glass breaking, sirens, alarms, and baby cries.

SoundID's on-device robust applications range from smart homes and security systems to healthcare and industrial automation. Its unique two stage approach offers remarkable accuracy and performance for complex sound detection tasks in environments where cloud revalidation is not feasible for privacy or accessibility reasons. For example, in a hospital setting, embedded SoundID has the ability to provide immediate or near-instantaneous sound recognition while complying with healthcare privacy regulations, such as HIPAA in the United States.
In parallel, the newly introduced SoundID cloud-based model is a ground-breaking addition to the realm of sound analysis. Capable of detecting a whopping 460 different sound classes with remarkable precision, it revolutionizes the applications of sound analysis. Sensory's SoundID could be tailored to recognize specific sounds relevant to products like the hum of a refrigerator, the beep of a microwave, or the spin cycle of a washing machine.

Running on an innovative deep learning algorithm, the cloud-based SoundID model can analyze audio streams in real-time, identifying the prevalent sounds of interest and responding accordingly. Clients can opt for an always-on model or a lower-power event triggered model. Furthermore, the on-device SoundID can be used in tandem with the cloud-based system for sound verification or as a backup, maintaining a balanced and seamless approach to power consumption and cloud usage.

The application of these updates enables the enhancement of devices like hearing aids, which would be able to analyze a wider range of sounds in their environment, thus providing a more nuanced and enriched listening experience for users; think dynamic noise canceling. The improvements can also be applied to automotive settings, enabling precise detection of potential malfunctions or accidents based on irregular sounds, supporting preventive maintenance, and enhancing operational safety. For consumer electronics like home appliances SoundID may be used to diagnose issues or identify unusual alerts; saving both customer time and customer service support hours.

"We are thrilled to bring these advancements in sound detection technology to the market," said a CEO Todd Mozer. "The potential applications of these updates are wide-ranging and exciting. We believe these enhancements will significantly improve the capabilities of numerous industries, and we look forward to seeing the positive impacts they will have."

4701 Patrick Henry Dr Building 7,
Santa Clara, CA 95054, United States

Sensory Inc. creates a safer and superior UX through vision and voice technologies. Sensory technologies are widely deployed in consumer electronics applications including mobile phones, automotive, wearables, toys, IoT, PCs, medical products, and various home electronics. Sensory's product line includes TrulyHandsfree voice control, TrulySecure biometric authentication, TrulyNatural large vocabulary natural language embedded speech recognition and SensoryCloud.ai, a complete AI as a Service platform designed for processing voice and vision AI workloads in the cloud. Sensory's technologies have shipped in over three billion units of leading consumer products.

This release was published on openPR.