Skip to main content
POST
Create Persona
Create a persona based on an audio clip. Personas can be used to generate music with consistent vocal characteristics.
This endpoint returns the created persona directly (not a task ID). For vocal personas (persona_type: "vox"), first extract the vocal stem using Get Vox Stem.

Request Body

string
required
The audio clip ID to create persona from.
string
required
The name of the persona.
boolean
default:true
required
Whether the persona is publicly accessible.
string
A description of the persona.
string
The type of persona. Use "vox" for vocal personas.
string
The vox audio ID returned from Get Vox Stem. Required for vocal personas.
number
The start time (in seconds) of the vocal segment to use.
number
The end time (in seconds) of the vocal segment to use.
string
Style tags describing the persona (e.g., “soulful vocals, R&B, smooth harmonies”).
string
Base64 encoded image for the persona avatar (format: data:image/png;base64,...).

Response

The response contains the complete persona object with all details.
string
The created persona ID.
string
The persona name.
string
The persona description.
string
The type of persona (e.g., "vox").
boolean
Whether the persona is public.
object
The source clip information.
number
The vocal start time.
number
The vocal end time.

Example

Response Example

Workflow: Creating a Vocal Persona

For creating a vocal persona, follow these steps:
Once created, personas can be used in the Generate Music endpoint via the persona parameter to create music with consistent vocal characteristics.