Tts Voice Ai
AI多语言文字转语音工具,支持中文、英文、日语、韩语、粤语语音生成、配音、有声书及语音克隆。
MIT-0 · Free to use, modify, and redistribute. No attribution required.
⭐ 1 · 46 · 0 current installs · 0 all-time installs
MIT-0
Security Scan
OpenClaw
Benign
high confidencePurpose & Capability
Name/description (TTS, voice cloning) match the declared env var (MINIMAX_API_KEY), required binary (python3) and pip dependency (requests). The code calls MiniMax endpoints and maps voices/languages; these are expected for a TTS client.
Instruction Scope
SKILL.md instructs setting MINIMAX_API_KEY and running the provided tts.py with CLI flags. The instructions and code only perform language detection, voice selection, HTTP POSTs to MiniMax APIs, and write an audio output file—no broad file reads, system interrogation, or external endpoints outside the stated service.
Install Mechanism
No install spec or third-party downloads are present. It's an instruction-and-script package that relies on python3 and the requests library (declared). Nothing is fetched from arbitrary URLs or installed to system paths.
Credentials
Only MINIMAX_API_KEY is required and is used by the script to authenticate to the MiniMax API. No other credentials, secrets, or unrelated env vars are requested or referenced.
Persistence & Privilege
Skill is not always-on and does not request elevated persistence. It does not modify other skills or system-wide configs; it just runs as a standalone script when invoked.
Assessment
This package appears to be a straightforward TTS client for the MiniMax service. Before installing: (1) confirm you trust the MiniMax endpoints (api.minimaxi.com / api.minimax.io) and are comfortable providing your MINIMAX_API_KEY; (2) review the full tts.py file locally (it only uses the API key, makes POST requests, and writes an output file); (3) be aware of legal/privacy implications of voice cloning or synthesizing third-party voices. If you do not trust or recognize the MiniMax service, do not provide your API key.Like a lobster shell, security has layers — review code before you run it.
Current versionv1.1.0
Download ziplatest
License
MIT-0
Free to use, modify, and redistribute. No attribution required.
SKILL.md
AI TTS 语音合成工具
支持 MiniMax 国内版(api.minimaxi.com)和国际版(api.minimax.io) API。
快速开始
1. 设置 API Key
# 国内版 (sk-api-xxx 格式)
export MINIMAX_API_KEY="sk-api-xxx"
# 国际版
export MINIMAX_API_KEY="your-key"
2. 生成语音
# 自动选择最佳音色
python3 tts.py "你好世界"
# 指定语言/风格
python3 tts.py "Hello world" --language english --gender female
# 粤语推荐使用 language_boost
python3 tts.py "你好" --language cantonese --language-boost "Chinese,Yue"
# 指定音色
python3 tts.py "你好" --voice girlfriend_5_speech02_01
# 列出音色
python3 tts.py --list-voices
参数说明
| 参数 | 简写 | 说明 | 默认值 |
|---|---|---|---|
--text | 要转语音的文本 | (必填) | |
--voice | -v | 音色ID | 自动匹配 |
--language | -l | 语言 | auto |
--gender | -g | 性别 (male/female) | auto |
--age | 年龄 (young/middle/elder/child) | auto | |
--style | -s | 风格 | auto |
--language-boost | -b | 语言增强 | auto |
--model | -m | 模型 | speech-2.8-turbo |
--speed | 语速 (0.5-2.0) | 1.0 | |
--format | -f | 格式 (mp3/wav/flac) | mp3 |
--output | -o | 输出文件 | output.mp3 |
--api | API版本 (cn/int) | cn | |
--list-voices | 列出音色 | - |
Language Boost (语言增强)
--language-boost 参数用于增强对小语种和方言的识别能力:
| 语言 | Boost 值 |
|---|---|
| 普通话 | Chinese |
| 粤语 | Chinese,Yue |
| 英语 | English |
| 日语 | Japanese |
| 韩语 | Korean |
| 自动检测 | auto |
注意: 粤语必须使用 Chinese,Yue,不是 Cantonese!
# 粤语示例
python3 tts.py "你食咗未呀" --language cantonese --language-boost "Chinese,Yue"
# 英语示例
python3 tts.py "Hello world" --language english --language-boost English
黄金音色推荐
✨ 中文 (Mandarin)
| 音色ID | 说明 | 风格 |
|---|---|---|
| girlfriend_5_speech02_01 | 温柔治愈 | 女声 |
| girlfriend_1_speech02_01 | 甜美软萌 | 女声 |
| ttv-voice-2026011810595326-hxYkopxR | 活泼吐槽 | 女声 |
| ttv-voice-2026011806464526-IPLmlZ8C | 清爽阳光 | 男声 |
| ttv-voice-2026011910402426-5fSKtVmM | 幽默大叔 | 男声 |
🌐 English
| 音色ID | 说明 |
|---|---|
| english_voice_agent_ivc_female_nora | 自然女声 |
| english_voice_agent_ivc_male_julian | 自然男声 |
| english_voice_agent_ivc_female_maya | 年轻女声 |
🇭🇰 Cantonese 粤语
| 音色ID | 说明 |
|---|---|
| HK_Cantonese_female1 | 标准港女 |
| Cantonese_ProfessionalHost(M) | 专业男主持 |
🇹🇼 Taiwan 台湾国语
| 音色ID | 说明 |
|---|---|
| vc_wanwan_0303_01 | 台湾女生 |
🇯🇵 Japanese
| 音色ID | 说明 |
|---|---|
| Japanese_DecisivePrincess | 果断公主 |
| Japanese_IntellectualSenior | 睿智老人 |
🇰🇷 Korean
| 音色ID | 说明 |
|---|---|
| Korean_SweetGirl | 甜妹 |
使用示例
视频配音 (Dubbing)
# 中文女声配音
python3 tts.py "欢迎观看本期节目" --voice girlfriend_5_speech02_01 --output intro.mp3
# 英文配音
python3 tts.py "Welcome to our channel" --language english --gender female --output intro_en.mp3
粤语配音 (强烈建议使用 language_boost)
# 粤语女声
python3 tts.py "你食咗未呀" --language cantonese --language-boost "Chinese,Yue" --voice HK_Cantonese_female1
# 粤语男声
python3 tts.py "大家好" --language cantonese --language-boost "Chinese,Yue" --voice "Cantonese_ProfessionalHost(M)"
有声书/朗读
# 温柔女声朗读
python3 tts.py "从前有座山" --style gentle
# 故事讲述
python3 tts.py "这是一个有趣的故事" --voice ttv-voice-2026011810595326-hxYkopxR
语音克隆
# 使用你的克隆音色
python3 tts.py "这是我的声音" --voice jiachuan_0115
语言代码
| 代码 | 语言 |
|---|---|
| chinese | 普通话 |
| cantonese | 粤语 (推荐配合 language-boost) |
| english | 英语 |
| japanese | 日语 |
| korean | 韩语 |
| taiwan | 台湾国语 |
API 版本
--api cn: 国内版 (api.minimaxi.com)--api int: 国际版 (api.minimax.io)
Files
2 totalSelect a file
Select a file to preview.
Comments
Loading comments…
