VoiceVault sets new standard for voice biometric accuracy
VoiceVault, a U.S. based private voice biometric company, has released the next generation of its voice biometric speaker verification engine. The new release enhances the identity verification accuracy demanded by its clients within the healthcare and financial services industries.
The firm’s voice biometrics solutions rely on the uniqueness of each person’s voice since no two voices are exactly the same. VoiceVault solutions have been used for biometric identity verification over the phone, Web, or via mobile and smart device applications.
New tests have confirmed that VoiceVault voice biometric solutions have obtained a verifiable equal error rate (EER) of 0.1 percent compared to typical error rates of around two percent in other voice biometric deployments. Its false accepts rate was ranked at 0.01 percent, while its false reject rates were less than 5 percent.
VoiceVault notes that this level of accuracy was obtained from a real-world financial services application that authorizes high-value financial transactions on a smartphone, where voice biometrics is part of a four-factor security cocktail. The better error and false reject rates provides the high level security needed by the financial services and healthcare industries especially for reducing fraud. It offers user convenience as well and does away with people remembering passwords.
Nick Wise, CTO, said: “VoiceVault works with many large healthcare and financial services institutions that demand very high levels of security in their speaker verification solutions. With this release, VoiceVault continues to meet the challenges they set and defines the standard for what voice biometrics need to deliver in order to gain acceptance in demanding high security applications such as mobile banking.”
Do you think voice biometrics will eventually be widely used in mobile banking applications?