Voice biometric technology identifies the speaker utilising their unique vocal characteristics such as pitch, tone and rhythm of speech. This is similar to other well-known biometric technologies such as fingerprint and retina scanning – and equally secure – however, it is the only practical means of authenticating a speaker over a remote channel e.g. a phone or PC.
Voice Biometrics is different from speech recognition – it identifies and verifies an individual’s identity – it is not about interpretation of the speaker’s words. It, therefore, measures both physical and behavioural characteristics of a person’s voice.
There are – broadly – two types of authentication:
- Active Caller Authentication which supports automation and self-service
- Passive Caller Authentication which is agent-assisted
Active authentication requires the customer or caller to be actively involved in the process i.e. they need to enroll their voice into a voice bank (once off) which will serve as a template against which all future authentications will be mapped. In active authentication, the customer will be required to repeat a specific phrase and is, therefore, text dependent.
Passive authentication, on the other hand, happens passively using biometric technology that captures the speaker’s voice as they speak naturally (text independent) and identifies it against past recordings. The customer is unaware that their voice is being matched or authenticated.
Request More Information
Fill in your details below and we will contact you.