← 返回 Skills 市场
71
总下载
0
收藏
0
当前安装
1
版本数
在 OpenClaw 中安装
/install mj-windows-faster-whisper
功能描述
Local speech-to-text with the faster-whisper backend (CTranslate2). Use when transcribing audio locally, setting up the faster-whisper model cache, or replac...
安全使用建议
This skill appears to be what it says: a helper for running faster-whisper locally. Before installing or following the steps, consider that: (1) pip-installing packages and downloading models will fetch third-party code and files from the network — use a virtual environment and review packages if you are cautious; (2) model files can be large and require disk space; (3) if you need a private model you may need a Hugging Face token (the skill does not request any secrets by default); (4) the SKILL.md gives a Windows-style example path — adapt it to your OS and avoid hardcoding usernames; and (5) ensure you have permission to transcribe any audio you process. If those points are acceptable, the skill's instructions are proportionate to its purpose.
能力评估
Purpose & Capability
The skill is described as a local faster-whisper (CTranslate2) transcription helper and its instructions request exactly the things needed for that: python/ffmpeg availability, installing faster-whisper/ctranslate2/huggingface_hub, and downloading a CTranslate2-style model folder from Hugging Face.
Instruction Scope
Instructions are scoped to setting up and using a local model (convert OGG/Opus to WAV, load a local model folder, transcribe). They explicitly reference GitHub and Hugging Face for downloads — this requires network access and may require authentication for private models. The example local path uses a specific Windows username (C:\Users\joshu...), which is just an example but could confuse non-Windows users; otherwise the runtime instructions do not ask to read unrelated files or secrets.
Install Mechanism
This is an instruction-only skill (no install spec). It tells the user/agent to pip-install packages (faster-whisper, ctranslate2, huggingface_hub) and to download a model from Hugging Face. Installing packages from PyPI and downloading models are expected for this use case but do involve running third-party code and transferring model files from the network — normal for local ML tooling but worth noting.
Credentials
The skill declares no required environment variables or credentials and its instructions do not demand unrelated secrets. The only possible credential scenario is an optional Hugging Face token if the chosen model is private or rate-limited, which is consistent with the stated workflow.
Persistence & Privilege
The skill does not request always:true, does not modify system-wide configs in its instructions, and is instruction-only (no code writing or autonomous persistence). Its requested level of presence is proportionate.
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 mj-windows-faster-whisper - 安装完成后，直接呼叫该 Skill 的名称或使用
/mj-windows-faster-whisper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
v1.0.0
Windows-local faster-whisper setup with benchmark-cleaned workflow and local model paths.
元数据
常见问题
MJ Windows Faster Whisper 是什么？
Local speech-to-text with the faster-whisper backend (CTranslate2). Use when transcribing audio locally, setting up the faster-whisper model cache, or replac... 它是一个面向 Claude Code / OpenClaw 的 AI Agent Skill 插件，目前累计下载 71 次。
如何安装 MJ Windows Faster Whisper？
在 OpenClaw 或 Claude Code 对话框中运行命令「/install mj-windows-faster-whisper」即可一键安装，无需额外配置。
MJ Windows Faster Whisper 是免费的吗？
是的，MJ Windows Faster Whisper 完全免费，采用 MIT-0 许可证，可自由下载、安装和使用。
MJ Windows Faster Whisper 支持哪些平台？
MJ Windows Faster Whisper 跨平台运行，可在任意部署了 OpenClaw / Claude Code 的环境中使用（cross-platform）。
谁开发了 MJ Windows Faster Whisper？
由 magejosh（@magejosh）开发并维护，当前版本 v1.0.0。
推荐 Skills