Setup & Installation
Or with the ClawHub CLI, for registry-managed skill folders outside a full OpenClaw workspace:
Version History
- Updated description to highlight always-on, local API server capabilities for STT/TTS on Mac. - Simplified and clarified project overview in SKILL.md. - Removed the run_stt.py script from the project.
What This Skill Does
Runs an OpenAI-compatible STT and TTS API server locally on Apple Silicon Macs using the MLX framework. Installs and manages itself as a persistent LaunchAgent via Homebrew. Accepts audio or video files for transcription and text strings for speech synthesis.
Runs fully offline on Apple Silicon with an OpenAI-compatible API, so existing tools that target OpenAI audio endpoints work without code changes or data leaving the machine.
When to use it
- Transcribing local meeting recordings without sending audio to the cloud
- Generating spoken audio from text in automation pipelines
- Replacing OpenAI Whisper/TTS API calls with a local drop-in endpoint
- Converting video files to text transcripts on a Mac
- Running always-on voice processing for offline or privacy-sensitive workflows
Example Workflow
Here's how your AI assistant might use this skill in practice.
User asks: transcribe this interview recording at ./interview.mp4
- 1Calls run_stt.sh with the video path
- 2ffmpeg converts the file to wav internally
- 3Server sends audio to local mlx-community/glm-asr-nano-2512-8bit model
- 4Transcript text is returned to stdout
Plain text transcript of the interview
Requirements
Accounts, API keys, or tools you or your AI assistant may need to set up while using this skill.