Skip to main content
POST
create voice persona from audio (instant clone)
Clone Voice (POST /suno/v2/voices) is a Voice Persona API. It runs the same official flow as Init + Complete in one async task: upload → vox stem → phrase → TTS verify → create. You only send a public audio_url. The phrase / TTS step is internal — no second recording UI. Init + Complete is unchanged and still free.
Billed. Failed Clone Voice tasks should not be billed. See Persona overview.
On /generate send persona.persona_id + persona.is_voice: true. artist_clip_id is optional for Voice Persona (persona_id alone selects task vox). Same-account ById — stay on the creating account.

Request

Create task

Returns { "taskId": "..." }. Poll Get Task Status until status is success.

Task result (data when success)

Copy persona_id. artist_clip_id may also be present — optional on later /generate:

Use on /generate

Authorizations

Authorization
string
header
required

Bearer authentication header of the form Bearer <token>, where <token> is your auth token.

Body

application/json
audio_url
string<uri>
required

公网可下载的参考音频 URL,MP3/WAV,时长 10240 秒,建议 3060 秒单人干声

Example:

"https://example.com/voice.mp3"

name
string

音色名称;不传则由系统生成

Example:

"My Voice"

description
string

音色描述

is_public
boolean
default:false

是否公开到官方 Suno。默认 false。true 时该声线会出现在官方广场。

Example:

false

vocal_start_s
number

人声区间起点(秒)。不传则官方自动检测,与旧平台 voices 一致。

Example:

0

vocal_end_s
number

人声区间终点(秒)。不传则官方自动检测,与旧平台 voices 一致。

Example:

30

Response

201 - application/json
taskId
string
required

task id. Use this id to query task status.

Example:

"15c257ff-43f7-4678-bd41-202ad6b8488b"