API 文档

轻湖文字转语音服务接口说明

语音合成接口

POST /api/v1/tts

输入文本，返回二进制音频流。Content-Type: application/json

请求参数
参数类型必填说明
textstring是待合成文本（≤ 5000 字符）
backendstring否kokoro | edgetts | auto（默认 auto）
voicestring否声音标识（默认取各后端配置的默认声音）
response_formatstring否mp3 | wav | opus | flac（默认 mp3）
speedfloat否语速倍率 0.25~4.0，默认 1.0
languagestring否语言代码（如 zh-CN、en-US）
响应示例
HTTP/1.1 200 OK
Content-Type: audio/mpeg
Transfer-Encoding: chunked

<audio binary stream>
curl 示例
curl -X POST http://localhost:36099/api/v1/tts \
  -H "Content-Type: application/json" \
  -d '{"text":"你好世界","backend":"kokoro","voice":"zf_xiaoxiao","response_format":"mp3","speed":1.0}' \
  --output output.mp3
声音列表查询

GET /api/v1/tts/voices?backend=kokoro

查询指定后端的可用声音列表

响应示例
{
  "code": 200,
  "data": {
    "backend": "kokoro",
    "voices": [
      { "id": "af_heart", "name": "Heart (American English Female)", "language": "en-US", "gender": "Female" },
      { "id": "zf_xiaoxiao", "name": "Xiaoxiao (Mandarin Chinese Female)", "language": "zh-CN", "gender": "Female" }
    ]
  }
}
curl 示例
curl http://localhost:36099/api/v1/tts/voices?backend=kokoro
后端状态查询

GET /api/v1/tts/backends

查询可用 TTS 后端及健康状态

响应示例
{
  "code": 200,
  "data": {
    "default_backend": "kokoro",
    "backends": [
      { "name": "kokoro", "available": true },
      { "name": "edgetts", "available": true }
    ]
  }
}
curl 示例
curl http://localhost:36099/api/v1/tts/backends
支持的音频格式
MP3 WAV OPUS FLAC

文本长度限制：5000 字符 · 语速范围：0.25~4.0