Sensory leverages voice, facial biometrics in TrulySecure 2.0 on-device authentication solution

Silicon Valley-based Sensory last week announced the availability of TrulySecure 2.0, a multimodal biometric authentication technology that leverages both voice and facial biometrics technologies to ensure secure access to application on mobile Android devices.

Sensory has built its reputation in the speech-recognition arena, but TrulySecure 2.0—available via the AppLock by Sensory application for Android mobile devices—also utilizes facial-recognition and deep-learning-neural-network technologies to improve security and accuracy, according to Sensory CEO Todd Mozer.

“Over the past year, we’ve substantially improved both our speaker-verification and our vision technology, and TrulySecure 2.0 is the result of that, which we think is really is the best of all worlds,” Mozer said during an interview with IWCE’s Urgent Communications. “It’s got a whole lot of convenience. It’s really fast at identifying the right user.

“You can use face, or you can use voice. Or, you can set it up in a mode to require face and voice, and that’s what we call the TrulySecure mode, because that’s the mode where we can get one in a million false accepts. So, it’s a very, very accurate solution, yet it’s still quite convenient and flexible.”

Other biometric approaches exist for user verification, but many of the most-discussed technologies—fingerprint and retinal scans—require additional hardware that often is not embedded in a mobile device or require connectivity to a cloud-computing environment, Mozer said. In contrast, Sensory’s TrulySecure 2.0 technology in embedded on the device—no cloud connection is needed—and utilizes microphones and cameras that are virtually ubiquitous on modern mobile devices, he said.

 “We got into vision about two years ago, and we had the strategy of combining vision and voice together to offer a really secure biometric authentication that was not only secure but really convenient,” Mozer said.

“We saw a real opportunity, because everyone kept telling us that they were frustrated with fingerprint—the manufacturers didn’t want to pay so much for fingerprint readers, and the end users got frustrated, because they kept getting locked out of the fingerprint things. In the era where people are trying to make things smaller and smaller, having a dedicated sensor for a fingerprint made no sense to us. We really wanted to use the microphone and camera that were built into mobile phones.”