← 返回 Skills 市场
Qwen3-TTS + Feishu Voice
作者
shawnreality
· GitHub ↗
· v1.0.3
396
总下载
0
收藏
1
当前安装
4
版本数
在 OpenClaw 中安装
/install qwen3-tts-feishu
功能描述
使用 Qwen3-TTS 本地语音合成，将文字转为语音文件，并可通过飞书发送语音消息（语音气泡格式）。支持 Apple Silicon (MPS) 和 CUDA GPU，无需 API Key 即可本地合成。
安全使用建议
This package appears coherent with its stated purpose, but before installing: (1) run in a contained environment (virtualenv or container) because pip packages like qwen-tts and modelscope will execute code; (2) verify the model source (ModelScope or Hugging Face repo) and licensing; (3) limit and rotate FEISHU_APP_ID/FEISHU_APP_SECRET and grant the app only the minimum Feishu permissions (im:message:send_as_bot, im:resource); (4) be aware of the large (~4.2GB) model download and GPU/CPU resource needs; (5) inspect or audit any third-party packages if you require higher assurance.
功能分析
Type: OpenClaw Skill
Name: qwen3-tts-feishu
Version: 1.0.3
The skill bundle provides legitimate functionality for local text-to-speech synthesis using Qwen3-TTS and integration with the Feishu (Lark) messaging platform. The scripts (scripts/synthesize.py and scripts/send_voice_feishu.sh) perform their stated tasks using standard libraries and official Feishu API endpoints, handling sensitive credentials via environment variables as required by the OpenClaw specification.
能力评估
Purpose & Capability
The name/description (local Qwen3-TTS + Feishu voice) matches the required binaries (ffmpeg, sox, python3), the two shipped scripts (synthesize.py and send_voice_feishu.sh), and the two required env vars (FEISHU_APP_ID, FEISHU_APP_SECRET). All requested items are proportional to producing audio locally and uploading it to Feishu.
Instruction Scope
SKILL.md and the scripts only reference the TTS model, local audio files, and Feishu APIs. The send script reads only the provided WAV file and FEISHU_* env vars, converts to opus, obtains a tenant token from open.feishu.cn, uploads the file, and posts the audio message. There are no instructions to read unrelated system files, other credentials, or to transmit data to unexpected endpoints.
Install Mechanism
This is instruction-only (no packaged installer). It instructs pip install qwen-tts, modelscope, soundfile and to download a ~4.2GB model from ModelScope or Hugging Face. Those are expected for local TTS but imply executing third-party Python packages and large model files — review package provenance before install and use a virtualenv.
Credentials
Only FEISHU_APP_ID and FEISHU_APP_SECRET are required, which matches the Feishu API workflow shown. No other secrets or unrelated credentials are requested or read by the scripts.
Persistence & Privilege
always:false and no hooks that modify other skills or system-wide configs. The skill does not request persistent elevated privileges or write to other skills' configurations.
如何使用
- 确保已安装 OpenClaw（本地或 Docker 部署）
- 在对话框中输入安装命令：
/install qwen3-tts-feishu - 安装完成后，直接呼叫该 Skill 的名称或使用
/qwen3-tts-feishu触发 - 根据 Skill 的参数说明提供必要输入，即可获得结构化输出
版本历史
v1.0.3
fix: 修正 frontmatter 使用正确的 metadata.openclaw schema，registry 现在会正确显示 FEISHU_APP_ID / FEISHU_APP_SECRET 为必须环境变量
v1.0.2
security: 去掉隐式读取本地配置文件的 fallback，凭证现在仅通过 FEISHU_APP_ID / FEISHU_APP_SECRET 环境变量传入；metadata 补充必需/可选环境变量声明
v1.0.1
fix: 飞书语音消息进度条修复——content 中需手动传入 duration 字段（毫秒），否则进度条停在最右端不动
v1.0.0
首次发布：本地 Qwen3-TTS 语音合成 + 飞书语音气泡发送，支持 Apple Silicon MPS 和 CUDA
元数据
常见问题
Qwen3-TTS + Feishu Voice 是什么？
使用 Qwen3-TTS 本地语音合成，将文字转为语音文件，并可通过飞书发送语音消息（语音气泡格式）。支持 Apple Silicon (MPS) 和 CUDA GPU，无需 API Key 即可本地合成。 它是一个面向 Claude Code / OpenClaw 的 AI Agent Skill 插件，目前累计下载 396 次。
如何安装 Qwen3-TTS + Feishu Voice？
在 OpenClaw 或 Claude Code 对话框中运行命令「/install qwen3-tts-feishu」即可一键安装，无需额外配置。
Qwen3-TTS + Feishu Voice 是免费的吗？
是的，Qwen3-TTS + Feishu Voice 完全免费（开源免费），可自由下载、安装和使用。
Qwen3-TTS + Feishu Voice 支持哪些平台？
Qwen3-TTS + Feishu Voice 跨平台运行，可在任意部署了 OpenClaw / Claude Code 的环境中使用（macos, linux）。
谁开发了 Qwen3-TTS + Feishu Voice？
由 shawnreality（@shawnreality）开发并维护，当前版本 v1.0.3。
推荐 Skills