ID R&D, a provider of AI-based voice and face biometrics, has integrated its technology with Vivoka, a French startup providing voice recognition technology. The use of ID R&D’s IDVoice technology in Vivoka’s Voice Development Kit is expected to enabling more development of applications that include voice biometric capabilities using edge computing.
Various consumer applications can benefit from personalization and authentication through use of voice biometrics. The biggest slice of consumer applications is in the smart home device market, which saw worldwide shipments of reach 801.5 million units in 2020, an increase of 4.5% over 2019, according to the International Data Corporation (IDC) Worldwide Quarterly Smart Home Device Tracker.
IDC cites convenience as a factor driving adoption of home automation and “ambient computing” products such as smart speakers. Shipments are expected to surpass 1.4 billion devices in 2025. Alexey Khitrov, ID R&D CEO, addressed the convenience factor.
“Biometrics has the unique ability to bring instant personalization and authentication to voice-enabled applications and devices of all kinds, Khitrov said. “It’s a natural fit and completely frictionless for users who don’t need to do anything other than speak their request in order for the system to quickly and accurately identify them.”
Speedy response of voice applications is steadily improving thanks to continual refinement of algorithms as well as advances in hardware and software for edge computing. “Running voice biometrics at the edge brings never-before realized value to smart speakers, smart cameras, connected cars, robotics and more,” Khitrov noted.
While edge computing in devices often rely on network connections back to a central cloud for data processing (data, in this case, being a person’s voice), Vivoka has taken a privacy-focused approach and focused on offline processing, meaning activity takes place on the device itself. ID R&D’s IDVoice can be used with ultra-short speech utterances and is text and language independent, the company said. Their focus on optimizing voice biometric technology to be both fast and small enough to run on chips such as Neural Processing Unit (NPU) that specialize in the acceleration of machine learning algorithms also aligns with Vivoka’s efforts on the application development front.
Vivoka voice development kit offers a graphical interface that makes it easier for developers to build applications by including a complete range of embedded voice technologies from multiple providers, the company said. Vivoka’s V3 VDK includes ID R&D voice biometric anti-spoofing for detecting attacks that use recorded or synthesized speech. These capabilities will prove useful in applications beyond the consumer market.
“As the voice market continues to evolve, new use cases for speech technology will benefit from faster and easier ways for developers and companies to build and test their ideas,” said William Simonin, CEO at Vivoka. “Our partnership with ID R&D provides access to state-of-the-art voice biometrics through our Voice Development Kit, opening up new possibilities for our customers to utilize speech technology to create better, more secure user experiences,” in markets such as robotics, mobility, aerospace, and other Industry 4.0 applications.
application development | ID R&D | Smart Devices | user authentication | Vivoka | voice biometrics | voice recognition Smart Home