Sensory Unveils TrulyNatural Voice Assistant SDK Specific for Smart Appliances

Sensory announced the release of a specialized model of TrulyNatural, the company's large vocabulary speech recognition and natural language understanding platform, that has been designed for voice at the edge. This domain specific assistant technology will help manufacturers create a new wave of smart kitchen appliances that don't relay voice requests to the cloud. Collaborating with Midea Microwave and Cleaner Appliances (MCA) business unit, Sensory will preview a microwave oven with a natural language user interface at CES 2020.

The Smart Audio Report Spring 2019 is part of the industry's longest-running public research series about smart speaker consumer behavior in the U.S.. The updated research reveals how smart speaker user behavior and perceptions are evolving as owners 'settle in' with their devices and the market matures out of its introductory phase. According to the survey promoted by NPR And Edison Research, seven-in-ten smart speaker owners use their device daily despite common concerns around security and privacy, while those same factors are leading reasons non-owners have not acquired a device.

The sound recognition market is the next growth area for software companies, semiconductor vendors, and those players that are currently dominating the voice recognition market, according to tech analysts SAR Insight & Consulting. After voice recognition become one of the greatest technological achievements in recent years, the market is ripe to leverage many of the same technologies and platforms, this time for smart devices being able to monitor sounds other than human voices, as the tech research firm states.

Speech and vision technologies specialist Sensory announced the second major update to TrulySecure, its FIDO Certified multimodal biometric authentication technology. TrulySecure quickly, conveniently and securely recognizes enrolled users through a fusion of face and voice biometric recognition technologies in less than a second, and differentiates individuals at a 99.999% accuracy rate – all without requiring any costly and space consuming biometric scanners.