Voice biometrics uses measurable characteristics of speech to verify or identify a person. A system compares a live or recorded sample with an enrolled voice profile and returns a match score. Because voices change and can be replayed or cloned, the score should support authentication rather than serve as unquestioned identity proof.
How does voice biometric verification work?
Enrolment creates a template from approved voice samples. Later, the system compares a new sample and applies a threshold. Lower thresholds admit more genuine users but also more impostors, so teams must set them against the action’s risk. Speaker identification selects among identities, while verification checks one claimed identity.
Which controls matter most?
Treat templates and recordings as sensitive biometric data with explicit consent, access, retention, and deletion rules. Pair confidence gating with another authentication factor for consequential actions. Test replay, channel changes, illness, ageing, and voice cloning. Include the system in AI risk management because false accepts and false rejects affect different people in different ways.
Frequently asked questions
Is a voiceprint an audio recording?
A voiceprint is usually a derived biometric template, although systems may also retain source audio under separate controls.
Can voice biometrics detect a cloned voice?
Some systems include presentation attack detection, but cloned and replayed speech remain risks that require layered authentication.
No advertising or tracking cookies, and our visitor counts are anonymous. The Cal.com booking widget loads only if you allow it. Privacy Policy.
The page itself, anything our host sets to serve and secure it, and the anonymous visitor count. Always on, and none of it stores anything on your device.
The Cal.com booking widget. Left off, a booking link opens the booking page instead of a popup, so you can still book a call.