# IndexTTS Extension Speech Synthesis

> Audio Generation

Creates speech through the OpenAI-compatible `/v1/audio/speech`
endpoint while accepting the IndexTTS-specific extension fields
documented by UModelverse.

## Endpoint

`POST https://api.modelverse.cn/v1/audio/speech`

## Request Body

## Responses

- **200** — Synthesized speech audio.
- **400** — Invalid request parameters.
- **default** — Error response.

## OpenAPI Definition

```json
{
  "openapi": "3.1.0",
  "x-language": "en-US",
  "info": {
    "title": "ModelVerse IndexTeam IndexTTS Extension Speech API",
    "version": "1.0.0",
    "description": "Self-contained OpenAPI 3.1 schema for the UModelverse\n`/v1/audio/speech` request fields documented for IndexTeam/IndexTTS\nseries models, including the platform-specific extension fields.\n\nThis source describes extension parameters layered onto the\nOpenAI-compatible TTS request shape. Extension fields are only meaningful\nwhen calling IndexTeam/IndexTTS series models on UModelverse.\n"
  },
  "servers": [
    {
      "url": "https://api.modelverse.cn",
      "description": "ModelVerse API endpoint shown in the source examples."
    }
  ],
  "tags": [
    {
      "name": "IndexTeam IndexTTS Extend",
      "description": "IndexTeam/IndexTTS speech synthesis with UModelverse extension fields."
    }
  ],
  "security": [
    {
      "bearerAuth": []
    }
  ],
  "paths": {
    "/v1/audio/speech": {
      "post": {
        "tags": [
          "IndexTeam IndexTTS Extend"
        ],
        "operationId": "createIndexTeamIndexTTSExtendSpeech",
        "summary": "IndexTTS Extension Speech Synthesis",
        "description": "Creates speech through the OpenAI-compatible `/v1/audio/speech`\nendpoint while accepting the IndexTTS-specific extension fields\ndocumented by UModelverse.\n",
        "requestBody": {
          "required": true,
          "content": {
            "application/json": {
              "schema": {
                "$ref": "#/components/schemas/IndexTeamIndexTTSExtendSpeechRequest"
              },
              "examples": {
                "happyBuiltinVoice": {
                  "summary": "Built-in voice with emotion text",
                  "value": {
                    "model": "IndexTeam/IndexTTS-2",
                    "input": "你好，这是一段带有愉快情绪的语音示例。",
                    "voice": "jack_cheng",
                    "sample_rate": 24000,
                    "gain": 1,
                    "emo_control_method": 1,
                    "emo_weight": 0.8,
                    "emo_text": "愉快",
                    "interval_silence": 200,
                    "max_text_tokens_per_sentence": 120
                  }
                },
                "customVoiceWithRandomEmotion": {
                  "summary": "Custom voice with emotion-vector control",
                  "value": {
                    "model": "IndexTeam/IndexTTS-2",
                    "input": "你好，我是带情绪的自定义音色示例。",
                    "voice": "uspeech:xxxx-xxxx-xxxx-xxxx",
                    "emo_control_method": 2,
                    "emo_weight": 0.6,
                    "emo_random": true,
                    "interval_silence": 200,
                    "max_text_tokens_per_sentence": 120
                  }
                }
              }
            }
          }
        },
        "responses": {
          "200": {
            "description": "Synthesized speech audio.",
            "content": {
              "audio/wav": {
                "schema": {
                  "$ref": "#/components/schemas/IndexTeamIndexTTSExtendAudioResponse"
                },
                "examples": {
                  "builtinVoiceSpeech": {
                    "summary": "WAV audio for a built-in voice request",
                    "value": "{binary WAV audio bytes}"
                  },
                  "customVoiceSpeech": {
                    "summary": "WAV audio for a custom voice request",
                    "value": "{binary WAV audio bytes}"
                  }
                }
              },
              "application/octet-stream": {
                "schema": {
                  "$ref": "#/components/schemas/IndexTeamIndexTTSExtendAudioResponse"
                },
                "examples": {
                  "speechFile": {
                    "summary": "Synthesized audio file bytes",
                    "value": "{binary audio bytes}"
                  }
                }
              }
            }
          },
          "400": {
            "description": "Invalid request parameters.",
            "content": {
              "application/json": {
                "schema": {
                  "$ref": "#/components/schemas/IndexTeamIndexTTSExtendErrorResponse"
                },
                "examples": {
                  "invalidExtensionParameter": {
                    "$ref": "#/components/examples/IndexTeamIndexTTSExtendError"
                  }
                }
              }
            }
          },
          "default": {
            "description": "Error response.",
            "content": {
              "application/json": {
                "schema": {
                  "$ref": "#/components/schemas/IndexTeamIndexTTSExtendErrorResponse"
                },
                "examples": {
                  "error": {
                    "$ref": "#/components/examples/IndexTeamIndexTTSExtendError"
                  }
                }
              }
            }
          }
        }
      }
    }
  },
  "components": {
    "securitySchemes": {
      "bearerAuth": {
        "type": "http",
        "scheme": "bearer",
        "bearerFormat": "API key",
        "description": "ModelVerse API key sent as `Authorization: Bearer <MODELVERSE_API_KEY>`."
      }
    },
    "schemas": {
      "IndexTeamIndexTTSExtendSpeechRequest": {
        "type": "object",
        "additionalProperties": false,
        "required": [
          "model",
          "input",
          "voice"
        ],
        "properties": {
          "model": {
            "$ref": "#/components/schemas/IndexTeamIndexTTSExtendModel"
          },
          "input": {
            "type": "string",
            "minLength": 1,
            "description": "Text to synthesize.",
            "examples": [
              "你好，欢迎使用 Modelverse TTS。"
            ]
          },
          "voice": {
            "type": "string",
            "minLength": 1,
            "description": "Voice to use for synthesis. The source examples show the built-in\n`jack_cheng` voice and custom voice IDs shaped like `uspeech:xxxx`.\n",
            "examples": [
              "jack_cheng",
              "uspeech:xxxx-xxxx-xxxx-xxxx"
            ]
          },
          "response_format": {
            "type": "string",
            "minLength": 1,
            "description": "OpenAI-compatible base TTS field mentioned by the source document.\nConcrete response-format values are not enumerated in this source.\n"
          },
          "speed": {
            "type": "number",
            "format": "double",
            "minimum": 0.25,
            "maximum": 4,
            "default": 1,
            "description": "Speech playback speed. The documented range is 0.25 to 4."
          },
          "instructions": {
            "type": "string",
            "description": "OpenAI-compatible base TTS field mentioned by the source document.\nThis source does not define additional IndexTTS-specific semantics.\n"
          },
          "sample_rate": {
            "type": "integer",
            "default": 22050,
            "description": "Target audio sample rate. Supported concrete values are defined by\nthe provider; the source lists examples such as 16000, 22050, and\n24000.\n",
            "examples": [
              16000,
              22050,
              24000
            ]
          },
          "gain": {
            "type": "number",
            "format": "double",
            "minimum": 0,
            "maximum": 10,
            "default": 1,
            "description": "Output volume gain coefficient. The source recommends the range\n`(0, 10]` and notes that `0` mutes output.\n"
          },
          "emo_control_method": {
            "type": "integer",
            "enum": [
              0,
              1,
              2,
              3
            ],
            "default": 0,
            "description": "Emotion-control method identifier. `0` disables emotion control,\n`1` uses emotion audio, `2` uses an emotion vector, and `3` uses\nemotion text.\n"
          },
          "emo_weight": {
            "type": "number",
            "format": "double",
            "minimum": 0,
            "maximum": 1,
            "default": 1,
            "description": "Weight applied to the emotion reference audio, vector, or text.\nThe documented valid range is 0.0 to 1.0. The source recommends\nabout 0.6 or lower for text emotion mode to produce more natural\nspeech.\n"
          },
          "emo_vec": {
            "type": "array",
            "minItems": 8,
            "maxItems": 8,
            "default": [
              0,
              0,
              0,
              0,
              0,
              0,
              0,
              0
            ],
            "description": "Emotion vector ordered as happiness, anger, sadness, fear,\ndisgust, melancholy, surprise, and calm. Each dimension is\ndocumented as 0 to 1.2, and the sum of all dimensions must not\nexceed 1.5.\n",
            "items": {
              "type": "number",
              "format": "double",
              "minimum": 0,
              "maximum": 1.2
            },
            "examples": [
              [
                0.8,
                0,
                0,
                0,
                0,
                0,
                0.1,
                0
              ]
            ]
          },
          "emo_text": {
            "type": "string",
            "default": "",
            "description": "Natural-language emotion prompt, such as happy, calm, or excited.",
            "examples": [
              "愉快",
              "平静",
              "激动"
            ]
          },
          "emo_random": {
            "type": "boolean",
            "default": false,
            "description": "Whether to introduce randomness into emotion control for more\nvariety or to avoid identical emotion expression between sentences.\n"
          },
          "interval_silence": {
            "type": "integer",
            "minimum": 0,
            "default": 200,
            "description": "Controls interval silence between sentences. The field table and\ncurl examples use a millisecond integer with a recommended value\nof 200.\n",
            "examples": [
              200
            ]
          },
          "max_text_tokens_per_sentence": {
            "type": "integer",
            "minimum": 1,
            "default": 120,
            "description": "Maximum token count or length threshold for internal sentence\nsplitting in long-text synthesis. The source recommends 120.\n"
          }
        }
      },
      "IndexTeamIndexTTSExtendModel": {
        "type": "string",
        "pattern": "^IndexTeam/IndexTTS(?:-.+)?$",
        "description": "IndexTeam/IndexTTS series model name. The source examples use\n`IndexTeam/IndexTTS-2`.\n",
        "examples": [
          "IndexTeam/IndexTTS-2"
        ]
      },
      "IndexTeamIndexTTSExtendAudioResponse": {
        "type": "string",
        "format": "binary",
        "description": "Synthesized speech audio bytes written by the source examples to `.wav` files."
      },
      "IndexTeamIndexTTSExtendErrorResponse": {
        "type": "object",
        "additionalProperties": true,
        "required": [
          "error"
        ],
        "properties": {
          "error": {
            "$ref": "#/components/schemas/IndexTeamIndexTTSExtendError"
          },
          "request_id": {
            "type": "string",
            "description": "Request identifier, if available."
          }
        }
      },
      "IndexTeamIndexTTSExtendError": {
        "type": "object",
        "additionalProperties": true,
        "required": [
          "message",
          "type"
        ],
        "properties": {
          "message": {
            "type": "string",
            "description": "Error message."
          },
          "type": {
            "type": "string",
            "description": "Error type.",
            "examples": [
              "invalid_request_error"
            ]
          },
          "code": {
            "type": [
              "string",
              "null"
            ],
            "description": "Error code.",
            "examples": [
              "param_error"
            ]
          },
          "param": {
            "type": [
              "string",
              "null"
            ],
            "description": "Request parameter related to the error.",
            "examples": [
              "emo_weight"
            ]
          }
        }
      }
    },
    "examples": {
      "IndexTeamIndexTTSExtendError": {
        "summary": "Invalid extension parameter",
        "value": {
          "error": {
            "message": "Invalid param",
            "type": "invalid_request_error",
            "code": "param_error",
            "param": "emo_weight"
          },
          "request_id": "request_id"
        }
      }
    }
  }
}
```