IndexTTS2音频合成

POST

/v1/audio/speech

最后修改时间：2025-10-14 11:57:48

B站开源其自研语音生成大模型 IndexTTS-2.0，是首个支持精确时长控制（Duration Control）的自回归（Autoregressive）零样本文本转语音（TTS）系统。

请求参数

Header 参数

Body 参数application/json必填

示例

{
  "model": "IndexTeam/IndexTTS-2",
  "input": "女人低着头，抱着她的脖子，长发披散着，根本看不见脸，只有发丝不断往下滴水",
  "max_tokens": 2048,
  "voice": "IndexTeam/IndexTTS-2:anna",
  "response_format": "mp3",
  "sample_rate": 32000,
  "stream": false,
  "speed": 1,
  "gain": 0
}

请求示例代码

Shell

JavaScript

Java

Swift

PHP

Python

HTTP

Objective-C

Ruby

OCaml

Dart

curl --location --request POST 'https://api.aiznt.com/v1/audio/speech' \
--header 'Authorization: Bearer ' \
--header 'Content-Type: application/json' \
--data-raw '{
  "model": "IndexTeam/IndexTTS-2",
  "input": "女人低着头，抱着她的脖子，长发披散着，根本看不见脸，只有发丝不断往下滴水",
  "max_tokens": 2048,
  "voice": "IndexTeam/IndexTTS-2:anna",
  "response_format": "mp3",
  "sample_rate": 32000,
  "stream": false,
  "speed": 1,
  "gain": 0
}'

返回响应

🟢200成功

application/json

Body

示例

{"url":"https://cdn.aiznt.com/audio/2025/10/14/04c4e1bc-b876-4cb4-b7f5-fad07ee619ff.mp3"}

修改于 2025-10-14 11:57:48

MOSS-TTSD音频合成

CosyVoice2音频合成