v1

latestOpenAPI 3.1.02026-07-13154170539.0 KB

Voices

TTS Design

Create Voice from Design Preview

Creates a full voice from a generated preview voice id.

post/v0/assistant/voice/design/create

Request body

generated_voice_idstring required

The 'generated_voice_id' obtained from the '/design/preview'.

voice_namestring required

Name for the new voice.

voice_descriptionstring required

Description for the new voice.

audio_base_64string nullable

Base64 encoded audio sample from the selected voice preview. If provided, it's used for language detection.

media_typestring nullable

MIME type of the audio sample, e.g., 'audio/mpeg'. Assumed 'audio/mpeg' if sample is provided but this is omitted.

labelsobject nullable

Optional labels for ElevenLabs when creating the voice.

languagestring nullable

Language of the voice. If not provided, it will be auto-detected from the provided audio preview, or from the description if no audio is provided.

genderstring nullable

Gender of the voice.

Response

Successful Response

Example response

{
  "info": {
    "voice_id": "bf0a246a-8642-498a-9950-80c35e9276b5",
    "name": "English Woman Calm 1",
    "description": "Calm and relaxing voice of an english-speaking woman",
    "gender": "female",
    "language": "en",
    "provider": "cartesia",
    "is_preset": true
  }
}