API ReferenceInstant Voice Cloning

Create Voice Clone

Create a zero-shot cloned voice from a multipart audio upload (tag + audio) or a JSON body with a publicly reachable audioUrl (tag + audioUrl). Pass an optional displayName to name the cloned voice; when it is omitted or empty the tag value is used as the display name. Returns a requestId (along with responseCode / responseMessage). Poll GET /v1/speech/voice-clone-creation-status/{requestId} until status is COMPLETED or FAILED.

Headers

tokenstringOptional
api-keystringOptional

Request

This endpoint expects a multipart form containing a file.
audiofileRequired

Reference audio file (WAV, MP3, FLAC, ALAW, or ULAW). Max 40 MB. Sample rate must be at least 24 kHz.

tagstringRequired
Unique identifier for the cloned voice. Mandatory. Use it to reference the voice and for any bookkeeping or grouping purposes of your own.
displayNamestringOptional
Optional display name for the cloned voice. If omitted or empty, the tag value is used as the display name.

Response

Clone job accepted. Poll creation status with the returned requestId.
requestIdstring
Id used to poll creation status.
responseCodestringOptional

Result code for the operation (for example, SUCCESS).

responseMessagestringOptional

Human-readable result message.

Errors

400
Bad Request Error
403
Forbidden Error
500
Internal Server Error
503
Service Unavailable Error