AudioVoice

One stable voice identifier exposed by a mounted TTS model.

idId (string)required

Model-specific voice identifier.

nameName (string)required

Human-readable voice name.

modelModel (string)required

Mounted text-to-speech model id.

preferred_languagesstring[]

Ordered BCP 47 language tags for which the model card recommends this voice.

Default value: []

kindKind (string)

Voice source: a model-native preset or a bundled reference profile.

Possible values: [builtin, reference]

Default value: builtin

AudioVoice

{
  "id": "string",
  "name": "string",
  "model": "string",
  "preferred_languages": [
    "string"
  ],
  "kind": "builtin"
}