xiaomi-mimo-v2-tts

Generate speech audio (WAV) from text using Xiaomi MiMo TTS (mimo-v2-tts model). Supports preset voices (mimo_default, default_zh, default_en), style control...

Security Scan

This skill's code, instructions, and required credential are consistent with a TTS client that sends text to Xiaomi MiMo's TTS API and writes back a WAV file.

Purpose & Capability

Name/description (MiMo TTS) matches the required environment variable (MIMO_API_KEY), the documented API endpoint (api.xiaomimimo.com), and the included script which implements the TTS calls.

Instruction Scope

SKILL.md and scripts/tts.py are narrowly focused on sending text (with optional style/user context) to the MiMo API and saving returned audio. The instructions do not attempt to read unrelated files, environment variables, or system state, nor do they exfiltrate data to unexpected endpoints.

Install Mechanism

There is no install spec and the skill is instruction-only plus a small Python script. No external downloads or archival extraction are used.

Credentials

Only one credential (MIMO_API_KEY) is required and it is used directly as an API key header in the HTTP request — proportional and expected for this purpose.

Persistence & Privilege

always is false and the skill does not request persistent system-level privileges or modify other skills' configs. It only suggests storing the API key in OpenClaw config or setting an environment variable.

Assessment

This skill appears coherent and limited to calling Xiaomi's MiMo TTS API and saving audio. Before installing: (1) Verify you trust the MiMo service and its privacy policy because any text you synthesize will be sent to api.xiaomimimo.com; (2) Keep your MIMO_API_KEY secret and only grant it to skills/services you trust; (3) Confirm any costs or rate limits associated with the API; (4) If you want to audit behavior, review scripts/tts.py — it uses plain HTTPS POST and base64-decoding and writes a local file; (5) Note the SKILL.md suggests storing the key via OpenClaw config or MIMO_API_KEY env var and restarting the session after setting it.

MiMo TTS — Speech Synthesis

First-Time Setup

If MIMO_API_KEY is not configured, the skill cannot function. Guide the user:

1. Obtain an API key from Xiaomi MiMo (https://api.xiaomimimo.com).
2. Set it via OpenClaw config:

openclaw config set skills.entries.mimo-tts.apiKey "your-api-key-here"

Or set the environment variable MIMO_API_KEY directly. After configuring, the user should restart or start a new session.

Generate Speech

Use scripts/tts.py to synthesize text to audio:

python3 "{baseDir}/scripts/tts.py" "要合成的文本" -o output.wav

Options

Flag	Default	Description
-o	output.wav	Output file path
-v	mimo_default	Voice: mimo_default, default_zh, default_en
-s	(none)	Style tag, e.g. 开心, 东北话, 悄悄话, 孙悟空
-f	wav	Audio format
--user-msg	(none)	Optional user-role context to adjust tone
--api-key	env MIMO_API_KEY	API key override

Examples

# Basic
python3 "{baseDir}/scripts/tts.py" "你好，今天天气真好" -o hello.wav

# Dialect style
python3 "{baseDir}/scripts/tts.py" "哎呀妈呀，这天儿也忒冷了吧" -s "东北话" -o dongbei.wav

# English voice
python3 "{baseDir}/scripts/tts.py" "Hello, how are you today?" -v default_en -o hello_en.wav

# Emotion + speed
python3 "{baseDir}/scripts/tts.py" "明天就是周五了，真开心！" -s "开心 变快" -o happy.wav

Style & Audio Tags

• Place <style>style</style> at the beginning of text to set overall style.
• Use inline audio tags for fine control: (紧张), (小声), (语速加快), (深呼吸), (苦笑), (沉默片刻).
• Multiple styles: <style>开心 变快</style>text.
• Singing: <style>唱歌</style>lyrics.

Voices

Name	voice param
MiMo-默认	mimo_default
MiMo-中文女声	default_zh
MiMo-英文女声	default_en