Sensory, a provider of speech and vision technologies for consumer electronics, has updated TrulySecure, its multimodal biometric authentication technology that uses a fusion of face and voice biometric recognition technologies.
Version 3.0 of TrulySecure offers improved recognition of enrolled users, especially in challenging scenarios. Sensory upgraded TrulySecure's embedded deep learning algorithms to better recognize users when a small part of the user's face is obstructed by wearable accessories. Sensory also improved its liveness detection algorithms.
TrulySecure now has software development kits and authentication solutions for both Android and iOS. Sensory has ported its AppLock app UX into the TrulySecure SDK for Android and iOS devices, offering a streamlined reference UI for customers to use as-is or modify for use within their own apps. Additionally, with the latest TrulySecure SDK, Android applications can now take advantage of the added layer of security offered by Android Key Store encryption to further protect the biometric data while not in use by the system.
"As security technologies continue to evolve, so will identity thieves. We continually improve the security and accuracy offered by TrulySecure to ensure that users are presented with a consistently better user experience, and that their apps and data are safe from prying eyes. In the latest release of TrulySecure, we have updated our deep learning convolutional neural nets to improve accuracy more than 50 percent and substantially decrease the ability to spoof with photos or videos," said Todd Mozer, CEO of Sensory, in a statement.
"Multisensor biometric security technologies are advancing with the help of increasingly sophisticated artificial intelligence algorithms," noted Brad Russell, a research analyst at Parks Associates, in a statement. "Technologies like TrulySecure, that can merge face and voice data for user authentication while reducing false rejects and minimizing exposure to the cloud, will be positioned well to replace less reliable methods."