首页 › 龙虾技能列表 › Jarvis Tts — 技能工具

Jarvis Tts — 技能工具

v1.0.0

[自动翻译] Jarvis TTS text-to-speech using Microsoft edge-tts with afplay playback. Use when users request voice output, audio responses, or text-to-speech. Prov...

2· 558·2 当前·2 累计
by @e421083458 (niuyufu)·MIT-0
下载技能包
License
MIT-0
最后更新
2026/2/26
安全扫描
VirusTotal
无害
查看报告
OpenClaw
安全
high confidence
The skill's code and instructions match its stated purpose (local macOS TTS using edge-tts + afplay); it does not request unrelated credentials or unusual installs, though it sends text to Microsoft's TTS service (network/privacy consideration).
评估建议
This package is internally consistent with a macOS TTS helper. Before installing or using it: 1) confirm you are okay with sending the text you feed it to Microsoft’s TTS service (privacy concern); 2) install and vet the edge-tts pip package from PyPI (verify publisher/reviews) rather than running arbitrary installs; 3) note it requires macOS afplay (or modify for Linux/Windows); 4) run the scripts on non-sensitive text first and consider running in a sandbox if you have high security requiremen...
详细分析 ▾
✓ 用途与能力
Name/description, SKILL.md, and the two included scripts all describe generating TTS with edge-tts and playing via afplay. There are no unrelated env vars, binaries, or config paths requested; the requested functionality (macOS TTS) reasonably explains the included files and dependencies.
✓ 指令范围
Instructions and scripts are narrowly scoped: take input text, run edge-tts to write an MP3 to a temp file, play it with afplay, then delete the temp file. They do not read other system files, environment variables, or send data to any endpoints other than the Microsoft TTS service invoked by edge-tts. No open-ended 'gather whatever context you need' behavior is present.
✓ 安装机制
There is no install spec in the package (instruction-only install); the scripts rely on standard Python and a user-installed pip package (edge-tts). No downloads from untrusted URLs or archive extraction are present in the bundle.
ℹ 凭证需求
The skill requires no credentials or env vars. However, edge-tts performs network calls to Microsoft’s TTS endpoints, so text input will be transmitted off-device to a third party; that is proportionate for a cloud-based TTS service but has privacy implications the user should be aware of.
✓ 持久化与权限
The skill is not always-on, does not modify other skills or system-wide settings, and does not request persistent presence. It performs only per-invocation local actions (generate/play/cleanup).
安全有层次，运行前请审查代码。

License

MIT-0

可自由使用、修改和再分发，无需署名。

运行时依赖

无特殊依赖

版本

latestv1.0.02026/2/22

Initial release of Jarvis TTS. - Provides natural-sounding Chinese text-to-speech using Microsoft edge-tts. - Playback handled by macOS afplay for smooth, uninterrupted audio. - Supports multiple Chinese voices (male/female). - Simple CLI script for fast conversion: jarvis-tts.sh "text". - Requires Python 3 and edge-tts package.

● 无害

安装命令 点击复制

官方npx clawhub@latest install jarvis-tts
镜像加速npx clawhub@latest install jarvis-tts --registry https://cn.clawhub-mirror.com

技能文档

功能

用 Microsoft edge-tts 生成高质量语音，然后用 macOS afplay 播放。提供自然流畅的中文语音输出。

使用场景

  • AI 助手语音回复
  • 文字转语音
  • 有声读物播放
  • 语音通知/提醒

优势

  • ✅ 自然 - 微软 Neural TTS，接近真人声音
  • ✅ 流畅 - 完整播放，不会中断
  • ✅ 多音色 - 支持多种中文语音（男声/女声）
  • ✅ 离线 - 生成后可重复播放

使用方法

基本用法

jarvis-tts.sh "要说的内容"

示例

# 简单回复
jarvis-tts.sh "好的，马上执行"
# 长文本
jarvis-tts.sh "从一数到一百：一、二、三...一百。数完了！"
# 指定语音
jarvis-tts.sh "你好" --voice zh-CN-YunxiNeural

可用语音

中文男声

  • zh-CN-YunxiNeural - 阳光活泼（默认）
  • zh-CN-YunjianNeural - 激情运动风
  • zh-CN-YunyangNeural - 专业新闻播报

中文女声

  • zh-CN-XiaoxiaoNeural - 温暖
  • zh-CN-XiaoyiNeural - 活泼

工作流程

文字输入 → edge-tts 生成 MP3 → afplay 播放 → 完成

详细步骤

  • 生成语音 - edge-tts 调用微软 TTS API 生成 MP3
  • 检查文件 - 确认生成成功且文件大小正常
  • 播放音频 - afplay 播放直到完成
  • 清理 - 删除临时文件

脚本说明

jarvis-tts.py

Python 脚本，执行 TTS 生成和播放。

依赖：

  • Python 3
  • edge-tts (pip3 install edge-tts)

用法： 

python3 jarvis-tts.py "要说的内容"

jarvis-tts.sh

Shell 封装脚本，方便直接调用。

用法： 

./jarvis-tts.sh "要说的内容"

技术细节

音频生成

python3 -m edge_tts \
  --voice zh-CN-YunxiNeural \
  --text "要说的内容" \
  --write-media /tmp/output.mp3

播放保证

  • 等待生成完成再播放
  • 检查文件大小确保成功
  • 同步播放直到完成

超时处理

  • 生成超时：60 秒
  • 播放超时：根据音频长度自动计算

限制

  • 仅支持 macOS（依赖 afplay）
  • 需要安装 edge-tts
  • 需要网络连接（调用微软 API）

扩展建议

如需支持其他平台：

  • Linux: 用 aplay 或 paplay 替代 afplay
  • Windows: 用 powershell -c (New-Object Media.SoundPlayer) 播放

相关文件

  • scripts/jarvis-tts.py - 主脚本
  • scripts/jarvis-tts.sh - Shell 封装
数据来源：ClawHub ↗ · 中文优化：龙虾技能库
OpenClaw 技能定制 / 插件定制 / 私有工作流定制

免费技能或插件可能存在安全风险，如需更匹配、更安全的方案，建议联系付费定制

了解定制服务