Voice cloning is AI that learns how a particular person sounds from a recording, then speaks any new text in that voice.
In plain words
A text-to-speech system normally reads in a stock voice. Voice cloning adds a sample of a real speaker, sometimes only seconds long, and the model copies that person's voice for new sentences. Some services clone instantly from a short clip; others train a custom model on more audio for higher accuracy. Because a cloned voice can be misused, providers such as Microsoft require the speaker's recorded consent.
A quick example
Imagine an app that lets a user read a consent statement and then a short sample aloud. The app creates their personal voice, and they can then have messages read out in it in many languages.
Where you'll meet it
- Personalised text to speech and narration.
- Dubbing and brand or character voices, with permission.
Related
Sources
- DocsWhat is personal voice for text to speech?, Microsoft Learn · read 29 Sept 2026
- DocsVoice cloning overview, ElevenLabs docs · read 29 Sept 2026
- PaperNeural Codec Language Models are Zero-Shot Text to Speech Synthesizers (VALL-E), Wang et al. (arXiv) · read 29 Sept 2026
Checked by WhatIsAI against the sources above · 29 Sept 2026. Spotted an error? Report it.