Phonexia DNN voice biometric engine authenticates callers in 3 seconds
Phonexia has released its latest generation of deep neural network (DNN)-powered Phonexia speaker identification technology, a biometric voice solution that authenticates callers in three seconds, the company announced.
Contact centers no longer have to rely on security questions to authenticate callers, Phonexia claims, as they can now leverage voice biometrics for enhanced accuracy and shorter verification times.
While the company’s previous speaker identification technology took 10 and even 30 seconds to identify a caller in 2019 and 2015 respectively, the upgraded engine provides accuracy above 90 percent within three seconds. Phonexia says customer-provided data can help further improve the speaker identification technology.
Phonexia Voice Verify for contact centers integrates the latest generation of Phonexia speaker identification technology to accurately identify clients over the phone based on voice biometrics in three seconds of net speech. The solution would help reduce contact center costs by skipping tens of seconds during the authentication phase, identify fraudsters and improve customer experience, and boost security, Phonexia says.
Last March, Phonexia released the fourth generation of its voice biometric software, which cuts the time required for identification by a third, while scoring accuracy above 99 percent in testing against the NIST SRE dataset.
The voice biometrics market is expected to grow at a CAGR higher than 23 percent to surpass $3.9 billion by 2026, driven by increasingly complex attacks.