> ## Documentation Index
> Fetch the complete documentation index at: https://docs.jimmyai.cn/llms.txt
> Use this file to discover all available pages before exploring further.

# SoundClone - Create Preview Task

> Submit a voice-cloning preview task. Poll GET /api/open-api/v1/audios/{id} for modelId and preview audioUrl.

<Note>
  SoundClone tasks are **async**. The create response returns a task `id`. Poll [Query SoundClone Task](/en/api-reference/common/sound-clone-query) until completion to obtain `modelId` and preview `audioUrl`, then call [Create Audio Task](/en/api-reference/sound-clone/audio-create) for production audio.

  All responses use the envelope `{ "code": 20000, "msg": "ok", "data": { ... } }`. Examples below show the `data` payload.
</Note>

## Overview

Submit a voice-cloning preview task from a source audio or video URL. When the task completes, you receive a preview audio URL and a `modelId` for formal audio generation.

## Request body

| Field          | Type   | Required | Description                                                                                                                                                                                                            |
| -------------- | ------ | -------- | ---------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------- |
| `fileUrl`      | string | Yes      | Public URL of source audio or video. No local paths or Chinese characters in the URL. Audio: `mp3`, `ogg`, `wav`, `m4a`, `aac`. Video: `mp4`, `avi`, `mov`, `mkv`, `flv`. Spoken content must be **> 15s and \< 60s**. |
| `contentText`  | string | No       | Preview script, max **270** characters. Uses a default line when omitted.                                                                                                                                              |
| `soundVersion` | string | No       | `v1` (24 languages) or `v2` (40 languages). Default `v1`.                                                                                                                                                              |
| `language`     | string | No       | Language code, default `auto`. Example: `Chinese`, `English`. Some languages require `v2`; see [Create Audio Task](/en/api-reference/sound-clone/audio-create).                                                        |

## Billing

Preview is billed **by character count** in units of **10,000 characters** (`price_mode`: `per_10k_char`).

| Model config          | Description                                             |
| --------------------- | ------------------------------------------------------- |
| `sound-cloning-clone` | Preview character fee; unit price is per 10k characters |

* Characters are counted as Unicode runes; `<#x#>` pause markers are excluded.
* The default preview text counts toward billing when `contentText` is omitted.
* Balance is checked before submission; failed tasks are refunded.

## Example

```bash theme={null}
curl --request POST \
  --url 'https://www.jimmyai.cn/api/open-api/v1/soundCloning/clones' \
  --header 'Authorization: Bearer sk_xxx' \
  --header 'Content-Type: application/json' \
  --data '{
    "fileUrl": "https://example.com/source-audio.mp3",
    "contentText": "A short preview sentence for voice cloning.",
    "soundVersion": "v1",
    "language": "Chinese"
  }'
```

## Response example

```json theme={null}
{
  "code": 20000,
  "msg": "ok",
  "data": {
    "id": "audio_16b635ba-5889-4fa5-bbcc-bf67a38c353a",
    "object": "audio",
    "created": 1781777280,
    "model": "soundCloningClone",
    "status": "queued",
    "error": null
  }
}
```


## OpenAPI

````yaml en/api-reference/openapi.json POST /api/open-api/v1/soundCloning/clones
openapi: 3.1.0
info:
  title: Jimmy AI OpenAPI
  description: API for Jimmy AI video generation services
  version: 1.0.0
servers:
  - url: https://www.jimmyai.cn
    description: Production server
security:
  - bearerAuth: []
paths:
  /api/open-api/v1/soundCloning/clones:
    post:
      summary: SoundClone - Create Preview Task
      description: >-
        Submit a voice-cloning preview task. Poll GET
        /api/open-api/v1/audios/{id} for modelId and preview audioUrl.
      operationId: createSoundClonePreview
      requestBody:
        required: true
        content:
          application/json:
            schema:
              $ref: '#/components/schemas/SoundCloneCloneRequest'
            examples:
              preview:
                summary: Create preview
                value:
                  fileUrl: https://example.com/source-audio.mp3
                  contentText: A short preview sentence for voice cloning.
                  soundVersion: v1
                  language: Chinese
      responses:
        '200':
          description: Task created
          content:
            application/json:
              schema:
                $ref: '#/components/schemas/SoundCloneApiResponse'
              examples:
                queued:
                  summary: Created
                  value:
                    code: 20000
                    msg: ok
                    data:
                      id: audio_16b635ba-5889-4fa5-bbcc-bf67a38c353a
                      object: audio
                      created: 1781777280
                      model: soundCloningClone
                      status: queued
                      error: null
components:
  schemas:
    SoundCloneCloneRequest:
      type: object
      required:
        - fileUrl
      properties:
        fileUrl:
          type: string
          description: Public URL of source audio or video
        contentText:
          type: string
          description: Preview script, max 270 characters
        soundVersion:
          type: string
          enum:
            - v1
            - v2
          default: v1
        language:
          type: string
          default: auto
      example:
        fileUrl: https://example.com/source-audio.mp3
        contentText: A short preview sentence for voice cloning.
        soundVersion: v1
        language: Chinese
    SoundCloneApiResponse:
      type: object
      properties:
        code:
          type: integer
          example: 20000
        msg:
          type: string
          example: ok
        data:
          $ref: '#/components/schemas/SoundCloneTaskData'
    SoundCloneTaskData:
      type: object
      properties:
        id:
          type: string
        object:
          type: string
          example: audio
        created:
          type: integer
          format: int64
        model:
          type: string
          enum:
            - soundCloningClone
            - soundCloningAudio
        status:
          type: string
          enum:
            - queued
            - processing
            - completed
            - failed
        error:
          type:
            - string
            - 'null'
      example:
        id: audio_16b635ba-5889-4fa5-bbcc-bf67a38c353a
        object: audio
        created: 1781777280
        model: soundCloningClone
        status: queued
        error: null
  securitySchemes:
    bearerAuth:
      type: http
      scheme: bearer

````