Speech Synthesis (Text to Audio)

High-quality speech synthesis interface supporting multiple voice tones, speed adjustment, and emotional expression. Suitable for audiobooks, voice assistants, and more.

API Endpoints

POST/t2a_v2

Text to speech

Request Parameters

Parameter	Type	Required	Description
`model`	string	Required	Model name: speech-2.6-hd or speech-2.6-turbo
`text`	string	Required	Text content to synthesize
`voice_id`	string	Optional	Voice ID, use preset or cloned voice
`speed`	number	Optional	Speech rate, range 0.5-2.0, default 1.0
`pitch`	number	Optional	Pitch, range -12 to 12
`audio_format`	string	Optional	Output format: mp3, wav, pcm

Request Example

{
  "model": "speech-2.6-hd",
  "text": "你好，欢迎使用 MiniMax 语音合成服务。",
  "voice_id": "female-tianmei",
  "speed": 1.0,
  "audio_format": "mp3"
}

Response Example

{
  "audio": "base64_encoded_audio_data...",
  "status": "success",
  "usage": {
    "characters": 18
  }
}

Code Examples

import requests
import base64

url = "https://your-proxy-domain.com/v1/t2a_v2"
headers = {
    "Authorization": "Bearer YOUR_API_KEY",
    "Content-Type": "application/json"
}
data = {
    "model": "speech-2.6-hd",
    "text": "你好，欢迎使用 MiniMax 语音合成服务。",
    "voice_id": "female-tianmei",
    "speed": 1.0,
    "audio_format": "mp3"
}

response = requests.post(url, headers=headers, json=data)
result = response.json()

# 保存音频文件
audio_data = base64.b64decode(result["audio"])
with open("output.mp3", "wb") as f:
    f.write(audio_data)