变声 / 音色转换

变声（音色转换）= 给一段原始音频（要保留的说话内容）+ 一段参考音频（目标音色），生成「用目标音色说出原内容」的语音。异步任务，提交后轮询查状态。

`POST /api/v2/voice/convert`

鉴权：Authorization: Bearer sk_...

字段	类型	必填	说明
`source_audio_url`	string	✅	原始说话音频 URL
`reference_audio_url`	string	✅	目标音色参考 URL
`diffusion_steps`	int	⬜	推理步数，越大质量越好越慢
`length_adjust`	number	⬜	时长缩放
`inference_cfg_rate`	number	⬜	CFG 强度
`auto_f0_adjust`	bool	⬜	自动调音高
`pitch_shift`	int	⬜	半音偏移
`return_format`	string	⬜	`wav` / `mp3`
`enable_separation`	bool	⬜	是否先做人声分离
`vocals_gain`	number	⬜	人声增益
`accompaniment_gain`	number	⬜	伴奏增益
`metadata`	object	⬜	透传业务字段

curl -X POST https://api.aiclonevoicefree.com/api/v2/voice/convert \
  -H "Authorization: Bearer sk_your_api_key" \
  -H "Content-Type: application/json" \
  -d '{
    "source_audio_url": "https://oss.aiclonevoicefree.com/noise_removal/1751766508261_mandarin_speech_16kHz.wav",
    "reference_audio_url": "https://oss.aiclonevoicefree.com/trump.wav",
    "return_format": "wav"
  }'

响应 202 Accepted

{
  "task_id": "99ae1ea9-01e3-4b15-9f89-d35c280f44b1",
  "status": "pending",
  "capability": "voice",
  "action": "vocal-conversion",
  "model": "voice-convert",
  "created_at": 1780704377
}

取结果

这是异步任务：上面拿到 task_id 后，轮询 GET /api/v2/tasks/{task_id}， status 变 completed 时 result 含转换后的音频：

curl https://api.aiclonevoicefree.com/api/v2/tasks/99ae1ea9-01e3-4b15-9f89-d35c280f44b1 \
  -H "Authorization: Bearer sk_your_api_key"

{
  "status": "completed",
  "capability": "voice",
  "action": "vocal-conversion",
  "audioUrl": "https://oss.aiclonevoicefree.com/...converted.wav",
  "format": "wav",
  "degraded": false,
  "providerUsed": "voice-convert-1",
  "completed_at": 1780704999
}

轮询节奏、状态枚举、统一响应格式详见任务查询。

计费

按输出音频时长计费：2 积分/秒（时长向上取整，最低 1 秒）。

算例：

输出约 30 秒 = 30 × 2 = 60 积分
输出约 12.4 秒 = 向上取整到 13 秒 = 13 × 2 = 26 积分

完成结算、失败不扣费（见通用约定）。用的是语音积分。

POST /api/v2/voice/convert

取结果

计费

On this page

`POST /api/v2/voice/convert`