theam/claude-open-marketplace/transcribe
Claude Code and agent skill by theam/claude-open-marketplace
- community
Checking that the SKILL.md file is still in the repository.
Transcribe audio files with speaker diarization using local ML models. Works in both Claude Code and Cowork mode.
Install
npx vskill@latest install theam/claude-open-marketplace/transcribeSKILL.md
Transcribe Audio
Transcribe an audio file locally with speaker diarization. All processing happens on your machine — no data leaves your device.
How to use
Run the CLI transcription tool via Bash. The plugin directory is:
Use the following command pattern:
Where {PLUGIN_DIR} is the absolute path to the transcription plugin directory (the directory containing transcribe_cli.py). To find it, look for the transcription plugin in the installed plugins — it will be under plugins/transcription/ in the monkey-tools plugin directory.
Arguments
| Argument | Description | |
|---|---|---|
audio_file | (required) Path to the audio file (.m4a, .mp3, .wav, .flac, .ogg, .aac, .mp4) | |
--language LANG | Language code (e.g., en, es). Auto-detected if omitted. | |
--skip-diarization | Skip speaker identification for faster processing. | |
--num-speakers N | Exact number of speakers if known. | |
--min-speakers N | Minimum expected number of speakers. | |
--max-speakers N | Maximum expected number of speakers. | |
--model REPO | MLX model override (e.g., mlx-community/whisper-large-v3-turbo for speed). | |
--export FORMAT | Output format: txt (default), json, or srt. |
Examples
Basic transcription:
With language hint and JSON export:
Fast mode (skip diarization):
Notes
- First run installs ML dependencies (~2GB). This is cached for subsequent uses.
- Apple Silicon: Automatically uses MLX for ~5x real-time transcription speed.
- Output: Summary stats go to stderr. Clean transcript goes to stdout.
- The transcript text is printed to stdout — capture it or read it directly from the Bash output.
Platform Notes
- Claude Code (macOS/Apple Silicon): Full support. Uses MLX GPU acceleration. Deps auto-install via
uvorpip. - Claude Code (Linux/Intel Mac): CPU-only mode (no MLX). Slower but functional.
- Cowork: CPU-only. Requires Python 3.11+ in the VM. Network restrictions may prevent auto-install of ML deps — pre-install with
uv pip install -e ".[ml]"if needed. - Pre-install deps to avoid first-run delay:
uv pip install -e ".[ml-apple]"(Apple Silicon) oruv pip install -e ".[ml]"(other platforms).
Security
- Security scanPassed, 90 of 100Pattern checks on the skill's files.
- Model reviewPassed
Quality evaluation
- Task evaluationNot runNo evaluation yet. Run one in Skill Studio.
Source
- SKILL.md in the repositoryCheckingplugins/transcription/skills/transcribe/SKILL.md
Popularity
- 7-day trend score
- 5
Not refreshed yet.
Works with
Any agent vskill installs to: 56 targets in all. These 39 have a page on this site.
- Claude Code
- Cursor
- GitHub Copilot
- Windsurf
- Codex
- Gemini CLI
- Cline
- Amp
- Roo Code
- Goose
- Continue
- Kilo Code
- Junie
- Trae
- OpenHands
- Qwen Code
- AI Agent
- Tabnine Chat
- Sourcegraph Cody
- Amazon Q
- Firebase Genkit
- Sweep
- Devin
- Phind
- Mentat
- Claude Desktop
- Aider
- GPT Pilot
- Plandex
- bolt.new
- v0
- Replit Agent
- Anse
- TypingMind
- BetterChatGPT
- LibreChat
- Warp AI
- Pieces
- Fig AI
Embed badge
[](https://verified-skill.com/skills/theam/claude-open-marketplace/transcribe)