API-driven advanced speech and signal processing technologies provider Aculab has announced the latest version of VoiSentry, its voice biometrics system.
The latest release includes the introduction of speaker identification capabilities for a range of uses including fraud detection, nuisance call interception, as well as for authentication for frictionless access. VoiSentry’s presentation attack detection (PAD) has also received an update to now include the ability to categorize suspected presentation attacks in line with FIDO specifications and according to the technology used.
“Regardless of the customer interaction channel, users want to present an unobtrusive and intuitive customer experience without having to compromise on security,” said Aculab Founder and CEO, Alan Pound. “At the same time, across all market sectors, businesses have to compete in terms of that experience, while fulfilling an overarching obligation to provide the highest levels of security of access to personal accounts and services,” he added.
VoiSentry — which is built upon a combination of machine learning, artificial intelligence, advanced Digital Signal Processing (DSP), and a variety of proprietary algorithms — also boasts improved accuracy via the continued development of its algorithms, along with the incorporation of the latest voice biometrics tech, a new Docker Container deployment option for Linux hosts, and a new API call that checks if audio recordings have originated from the same utterance.
“VoiSentry is designed as a scalable voice biometric system that is simple to install and integrate, with several deployment options in terms of platform environment,” explains Ian Colville, Product Manager at Aculab. “Control of Aculab’s system is via the VoiSentry API, which facilitates ready integration with any IVR, self-service, workforce management, or contact centre solution.”
April 27, 2020 – by Tony Bitzionis