Text-Independent Voice Biometrics

Last Updated
27 Feb 2021

A form of Voice Biometrics that is not dependent on the utterance being used during enrolment being repeated during authentication or identification.

Most often used alongside agent conversations with a Passive Voice Biometrics approach. The biometric algorithm is optimised for how the speaker generally speaks. As a result, it often requires longer utterances for enrolment during registration and authentication/identification to achieve the same level of performance as Text-Dependent Voice Biometrics. The gap is, however, closing as the technology improves. The major advantage is that these utterances can be taken from normal conversations.

Related Terms

Popular Posts