Skip to main content
Character Voices does for a voice what Characters does for a face. A Character can carry a voice, and the video models that support voices use it when the Character speaks. Audio references are the related feature: an audio file you attach to a generation to steer the sound.

Which models support Character Voices

Which models take audio references

On Cherry, audio references and Character Voices share the same slots.

Three different things

Limitations

  • Consistency depends on the quality of what you upload. Clean samples produce a more stable voice than phone recordings with room noise.
  • Only use voices you have the right to use. Cloning someone’s voice without their consent is not permitted. See likeness, consent, and reported content.
Last modified on September 14, 2026