Skip to content

Multimodal AI

Speaker verification

Checking whether a voice sample matches a claimed speaker identity.

Example

A service compares a caller's voice with an enrolled voice reference.

Why people use it

It provides one possible check that a caller matches a claimed identity.

What you'll hear

“Does the caller sound like the enrolled account holder?”

What this means for you

Use appropriate additional checks for consequential access decisions.

Can you control it?

Sometimes

Sometimes. Your choices depend on the tool and your access. The settings available to an everyday user may differ from those available to the people running it.

Common questions

Can voice checking be fooled by imitation or generated audio?
It can be vulnerable; resistance depends on the system and attack conditions.
Is it the same as checking what the person says?
No. It compares voice characteristics rather than whether the spoken statement is true.
Can a genuine caller be rejected?
Yes. Illness, poor audio or a changed speaking style can affect the match.

Related terms

Still have questions?

Up to 500 characters.

Ask LATHIC about AI. Relevant glossary entries may be included.

Your question, the glossary entries it matches, and a rotating pseudonymous identifier go to Microsoft Azure’s OpenAI service through Vercel AI Gateway to generate an answer. Zero retention and no training are required of the provider, and LATHIC does not save your question or answer. Privacy Notice