Our voices are unique.  The great comedians of the world may disagree, but fundamentally, it is impossible for one human to accurately replicate the voice of another.  It may be pretty much the same, but either the physiological (the mouth) or behavioral (the accent) will prevent two voices from being inherantly identical.  Consequently, voice biometrics may be used to identify individuals.

On enrolment into a voice recognition biometric system, the individual will provide a passphrase, repeated several times until their voice is characterised and recorded.  This may be done in a variety of settings, but for a high accuracy it is recommended to use the highest quality audio input available.

One benefit of voice recognition was invented by Alexander Graham Bell - the telephone.  I don’t have facts to hand about how many households in the developed world have one, but I’m sure it’s the majority.  Voice biometrics can be applied here, and with the low perceived invasiveness, it is popular with the public.  Unfortunately, high false non-matching rates are coupled with this technology.

Voice biometric systems are mostly used for telephony-based applications. Voice verification is used for government, healthcare, call centers, electronic commerce, financial services, customer authentication for service calls, and for house arrest and probation-related authentication.