← Back to Skills Marketplace
xcchenx345

mimo-tts-wav

by xcchenx345 · GitHub ↗ · v1.0.0 · MIT-0
cross-platform ⚠ pending
35
Downloads
0
Stars
0
Active Installs
1
Versions
Install in OpenClaw
/install mimo-tts-wav
Description
小米 MiMo-V2.5-TTS 语音合成工具。将文本合成为 WAV 音频文件,支持预置音色、语音设计、音色克隆。使用 OpenAI SDK 调用 mimo-v2.5-tts / mimo-v2.5-tts-voicedesign / mimo-v2.5-tts-voiceclone 模型。需要 MIMO_API...
README (SKILL.md)

MiMo TTS WAV 语音合成

调用小米 MiMo-V2.5-TTS 系列的 OpenAI 兼容 API,将文本合成为 WAV 音频文件。

前置条件

  • 环境变量 MIMO_API_KEY 已设置(从 xiaomimimo.com 获取)
  • Python 依赖:openai, numpy, soundfile(如用流式)

使用方式

快速合成(非流式)

import os
from openai import OpenAI
import base64

client = OpenAI(
    api_key=os.environ["MIMO_API_KEY"],
    base_url="https://api.xiaomimimo.com/v1"
)

completion = client.chat.completions.create(
    model="mimo-v2.5-tts",
    messages=[
        {"role": "user", "content": "用轻快活泼的语气读出这段文字"},
        {"role": "assistant", "content": "你好,欢迎收听今天的节目。"}
    ],
    audio={
        "format": "wav",
        "voice": "冰糖"  # 预置音色
    }
)

audio_bytes = base64.b64decode(completion.choices[0].message.audio.data)
with open("output.wav", "wb") as f:
    f.write(audio_bytes)

可用模型

Model ID 功能
mimo-v2.5-tts 预置音色合成(支持唱歌)
mimo-v2.5-tts-voicedesign 文本描述定制音色
mimo-v2.5-tts-voiceclone 音频样本复刻音色

预置音色(mimo-v2.5-tts)

  • mimo_default(中国集群=冰糖,海外集群=Mia)
  • 冰糖(中文女声)
  • 茉莉(中文女声)
  • 苏打(中文男声)
  • 白桦(中文男声)
  • Mia(英文女声)
  • Chloe(英文女声)
  • Milo(英文男声)
  • Dean(英文男声)

风格控制

两种方式(二选一):

  1. 自然语言控制 → 写在 role: usercontent
  2. 音频标签控制 → 在 role: assistantcontent 中嵌入 (风格)[音频标签]

references/style-guide.md 获取详细风格标签列表。

脚本

  • scripts/tts_simple.py — 单次非流式合成 WAV
  • scripts/tts_with_style.py — 带风格控制的合成示例
  • scripts/tts_stream.py — 流式合成并拼接 WAV

运行方式:python scripts/tts_simple.py "要合成的文本"

Capability Tags
requires-sensitive-credentials
How to Use
  1. Make sure OpenClaw is installed (local or Docker)
  2. Run the install command in chat: /install mimo-tts-wav
  3. After installation, invoke the skill by name or use /mimo-tts-wav
  4. Provide required inputs per the skill's parameter spec and get structured output
Version History
v1.0.0
- Initial release of mimo-tts-wav. - Provides a tool for synthesizing text into WAV audio using Xiaomi MiMo-V2.5-TTS models. - Supports preset voices, voice design by text, and voice cloning. - Integrates with OpenAI-compatible API, requiring MIMO_API_KEY. - Includes scripts for single synthesis, style control, and streaming synthesis. - Suitable for generating voice-overs, audiobooks, and voice messages.
Metadata
Slug mimo-tts-wav
Version 1.0.0
License MIT-0
All-time Installs 0
Active Installs 0
Total Versions 1
Frequently Asked Questions

What is mimo-tts-wav?

小米 MiMo-V2.5-TTS 语音合成工具。将文本合成为 WAV 音频文件,支持预置音色、语音设计、音色克隆。使用 OpenAI SDK 调用 mimo-v2.5-tts / mimo-v2.5-tts-voicedesign / mimo-v2.5-tts-voiceclone 模型。需要 MIMO_API... It is an AI Agent Skill for Claude Code / OpenClaw, with 35 downloads so far.

How do I install mimo-tts-wav?

Run "/install mimo-tts-wav" in the OpenClaw or Claude Code chat to install it in one step — no extra setup required.

Is mimo-tts-wav free?

Yes, mimo-tts-wav is completely free, licensed under MIT-0. You can download, install and use it at no cost.

Which platforms does mimo-tts-wav support?

mimo-tts-wav is cross-platform and runs anywhere OpenClaw / Claude Code is available (cross-platform).

Who created mimo-tts-wav?

It is built and maintained by xcchenx345 (@xcchenx345); the current version is v1.0.0.

💬 Comments