ID R&D and Synaptics Deploy Voice Biometrics on NPU
ID R&D, a biometric and liveness detection technology company, has integrated with Synaptics on its VideoSmart VS600 platform that leverages SyNAP technology to enable edge computing devices to take advantage of voice biometrics for user identification, authentication, and personalization. The Synaptics and ID R&D solution works with an array of voice-enabled smart home devices, including set-top boxes, smart speakers, displays, and security systems.
The capability is the result of integrating ID R&D's voice biometrics and voice anti-spoofing onto the neural network processing unit (NPU) of the Synaptics VS600 multimedia system-on-a-chip (SoC) solution. Synaptics' proven and versatile NPU supports trillions of operations per second with on-device execution of deep-learning models. ID R&D's ability to tap into the NPU enables sophisticated algorithms that result in AI-based voice biometrics on an extremely small footprint.
Moving the inferencing component of the voice biometric capabilities off the CPU to the NPU results in more than a tenfold speed improvement, a reduction of CPU utilization by a factor of three, and a significant reduction in RAM requirements, according to both companies. Additionally, the integration includes voice liveness to detect spoofing attacks such as recordings or voice deepfakes.
"This is the first known commercial application of voice biometrics with liveness on an NPU for smart home applications. The speed of the voice biometric processing is faster than real time, meaning we can identify a speaker in a family group even before they are finished speaking a wake word and a command to a home-based smart device," said John Amein, senior vice president of ID R&D, in a statement. "Previous systems have been limited by the capacity of the core CPU, which is not optimized for neural networks. This integration makes it possible to implement more sophisticated algorithms, opening up new use cases. We expect to enable voice for higher security levels necessary for use cases like payments, highly responsive identification of a person among a group for personalized responses, and a more natural user experience for authentication."
"The ability to instantly and reliably recognize a specific user on a shared device enables seamless personalization of controls and content without the frustrations associated with remembering and typing passwords," said Vineet Ganju, vice president of marketing for edge computing at Synaptics, in a statement. "The solution also enables restricted access where needed, such as enabling parental controls."
ID R&D and SimpleTech have paired on in-app voice biometrics to secure WhatsApp payment transactions.
27 Jan 2021