Minimax tts

Skill aAAaqwq/AGI-Super-Team/skills/minimax-tts

Text-to-speech synthesis via MiniMax WebSocket APIFrom its SKILL.md

Install

npx -y skills add aAAaqwq/AGI-Super-Team --skill minimax-tts

Assembled from the repository path, not quoted from the project. Check it against their README if it does not work.

SKILL.md

1.9 KB, 718 tokens by cl100k_base, as published. Nobody here has run it

MiniMax TTS (WebSocket)

触发条件

"语音合成"、"TTS"、"文字转语音"、"语音生成"、"MiniMax TTS"

概述

通过 MiniMax WebSocket API (wss://api.minimaxi.com/ws/v1/t2a_v2) 进行语音合成。

支持模型

模型	特性
speech-2.8-hd	精准还原真实语气细节，全面提升音色相似度
speech-2.6-hd	超低延时，归一化升级，更高自然度
speech-2.8-turbo	精准还原真实语气细节，更快更优惠
speech-2.6-turbo	极速版，适用于语音聊天和数字人
speech-02-hd	出色韵律、稳定性和复刻相似度
speech-02-turbo	小语种能力加强

常用中文音色

Voice ID	名称
female-yujie	御姐
female-shaonv	少女
female-shaonv-jingpin	少女-beta
female-tianmei	甜美女性
qiaopi_mengmei	俏皮萌妹
wumei_yujie	妩媚御姐
male-qn-qingse	青涩青年
male-qn-jingying	精英青年
Cantonese_CuteGirl	可爱女孩(粤语)
Cantonese_GentleLady	温柔女声(粤语)

完整列表: https://platform.minimaxi.com/docs/faq/system-voice-id

API 密钥

Key 来源: openclaw.json → env.vars → MINIMAX_API_KEY
端点: wss://api.minimaxi.com/ws/v1/t2a_v2
认证: Authorization: Bearer {api_key}

使用方法

from tts_ws import synthesize

await synthesize(
    model="speech-2.8-hd",
    voice_id="female-yujie",
    text="你好世界",
    api_key="...",
    output_path="/path/to/output.mp3"
)

注意事项

依赖: websockets (pip3 install websockets)
SSL: 当前跳过证书验证 (ssl.CERT_NONE)
Clash TUN: api.minimaxi.com 已配置 DIRECT 规则，WebSocket 可正常连接
单次最长 10,000 字符
输出格式: mp3 (32000Hz, 128kbps)

What ships with it: 1 file

4.0 KB alongside SKILL.md, 1 of them executable

scripts/

tts_ws.pyruns4.0 KB