How Does Voice Biometrics Work?

In Voice Biometrics, an individual’s spoken utterance is captured by the Biometric system and compared with a previously stored voiceprint, and the result is assigned a score. The voiceprint is an algorithmically derived model of physical and behavioral characteristics of a speaker’s voice and its unique digital representation. It is not a recording; nor a set of words; nor a wave pattern of a voice. It cannot be played back or used for any other purpose than a comparison with subsequent voiceprints using the same algorithm. 

The algorithm uses more than 100 different calculations to generate a voiceprint. An adaptation process can be used to track changes to a speaker’s vocal tract/voice over time as they age, avoiding any need to continuously update and maintain a voiceprint ‘identity’.

Because the voiceprint is characterized by the vocal trait (a physiological trait) and not specifically the sound of the utterance, afflictions that affect the ‘sound’ of a voice; the common cold, for example, do not affect the vocal tract and don’t have an adverse affect on accuracy levels. If a speaker can provide enough vocal energy to be heard, enough of their voice will be present for verification.

2 comments: Leave Your Comments