Voice cloning
Voice cloning is the creation of a synthetic text-to-speech voice that imitates a specific real person, built from recordings of their speech.
Also called: custom voice, synthetic voice clone
Where standard TTS offers a catalog of voices, cloning produces a new one modeled on an individual, an owner who wants the phone answered in their own voice, for example. Depending on the technique it needs anywhere from a few seconds of audio to an hour, and the result can be very close to the original.
Because a cloned voice can say anything, the practice raises consent and impersonation questions that a stock voice does not. Reputable providers require verifiable consent from the person being cloned, and regulators have taken interest: in the United States the Federal Communications Commission has treated AI-generated voices in outbound calls as artificial voices under existing robocall rules.
For a business the practical questions are whether the person consents, whether callers should be told the voice is synthetic, and whether a familiar voice on an AI line helps or confuses.