CosyVoice

创建克隆音色（CosyVoice）

通过上传音频创建 CosyVoice 克隆音色。

POST

/services/audio/tts/customization

cURL

curl -X POST https://dashscope.aliyuncs.com/api/v1/services/audio/tts/customization \
-H "Authorization: Bearer $DASHSCOPE_API_KEY" \
-H "Content-Type: application/json" \
-d '{
  "model": "voice-enrollment",
  "input": {
    "action": "create_voice",
    "target_model": "cosyvoice-v3.5-plus",
    "prefix": "myvoice",
    "url": "https://your-audio-url.wav",
    "language_hints": ["zh"]
  }
}'

{
  "output": {
    "voice_id": "cosyvoice-v3.5-plus-myvoice-xxxxxx"
  },
  "usage": {
    "count": 1
  },
  "request_id": "xxxxxxxx-xxxx-xxxx-xxxx-xxxxxxxxxxxx"
}

音色创建后需要审核。通过查询音色详情确认状态为 OK 后方可使用。状态说明参见概述。

鉴权

string

header

必填

DashScope API Key。前往获取 API Key。

请求体

application/json

enum<string>

必填

固定为 voice-enrollment。

可选值：voice-enrollment

示例:voice-enrollment

object

必填

显示子属性

enum<string>

必填

固定为 create_voice。

可选值：create_voice

示例:create_voice

enum<string>

必填

克隆音色绑定的语音合成模型。后续合成调用的 model 必须与此一致。

可选值：cosyvoice-v3.5-plus,cosyvoice-v3.5-flash,cosyvoice-v3-plus,cosyvoice-v3-flash

示例:cosyvoice-v3.5-plus

string

必填

音色名称前缀，仅限字母和数字，最长 10 个字符。生成的名称格式：{target_model}-{prefix}-{unique_id}。

示例:myvoice

取值范围：length <= 10pattern: ^[a-zA-Z0-9]+$

string

必填

用于克隆的音频文件 URL，必须可公开访问。

示例:https://your-audio-url.wav

string[]

默认值["zh"]

音频的语言提示，仅使用第一个元素。默认：["zh"]。

示例:

[
  "zh"
]

number

默认值10

预处理后的最大音频时长（秒）。范围：[3.0, 30.0]。默认：10.0。

示例:10

取值范围：3 <= x <= 30

boolean

默认值false

是否启用音频预处理（降噪、增强、音量归一化）。默认：false。

示例:false

响应

200-application/json

object

显示子属性

string

生成的音色 ID。在合成调用中将此值作为 voice 参数传入。

示例:cosyvoice-v3.5-plus-myvoice-xxxxxx

object

显示子属性

integer

计费的声音创建次数。成功创建时固定为 1。

示例:1

string

请求 ID，用于问题排查。

示例:xxxxxxxx-xxxx-xxxx-xxxx-xxxxxxxxxxxx

cURL

curl -X POST https://dashscope.aliyuncs.com/api/v1/services/audio/tts/customization \
-H "Authorization: Bearer $DASHSCOPE_API_KEY" \
-H "Content-Type: application/json" \
-d '{
  "model": "voice-enrollment",
  "input": {
    "action": "create_voice",
    "target_model": "cosyvoice-v3.5-plus",
    "prefix": "myvoice",
    "url": "https://your-audio-url.wav",
    "language_hints": ["zh"]
  }
}'

{
  "output": {
    "voice_id": "cosyvoice-v3.5-plus-myvoice-xxxxxx"
  },
  "usage": {
    "count": 1
  },
  "request_id": "xxxxxxxx-xxxx-xxxx-xxxx-xxxxxxxxxxxx"
}