Voice cloning

Voice cloning makes a text-to-speech model speak in a particular person's voice, often from only a few seconds of recording.

  • Using models
  • Updated

Zero-shot cloning needs no training: you give the model a short clip, for example three seconds, and it imitates that voice for new text. Some models also offer voice design, where you describe a voice in words instead. Quality depends on the clip, and clean audio of a single speaker works best.

Cloning someone's voice without permission can break the law and enable fraud, so use only voices you have the right to use. Some projects add an inaudible watermark to their output, and licenses may forbid impersonation or limit a model to non-commercial use. Read both before you build a product on a cloning model.

Related terms

  • Text-to-speech (TTS) Text-to-speech models turn written text into spoken audio, and many can copy a voice from a short sample.
  • Non-commercial license A non-commercial license lets you use a model for research, study and personal projects, but not to make money from it.

Keep reading

All terms