Overview

You can either choose from our stock voices or create your own custom voices:

• Instant clones: voices created from seconds of source audio (zero-shot)
• Professional clones: voices created from a minimum of 5 minutes of source audio (fine-tuned)

You can view voices in the Playground or with a call to our API.

​

Creating custom voices

You can create a custom voice from source audio files through our Playground, our REST API, and SDKs.

​

Custom professional clones: getting the best results

• Source material should:
  • Contain only one person’s voice, in English
  • Not contain background noise or music
• Consider Adobe’s speech enhancer if your speech is not professionally recorded
• If cutting down larger source material:
  • Leave gaps between sentences
  • Or upload sections as separate files