Setup & Installation
Or with the ClawHub CLI, for registry-managed skill folders outside a full OpenClaw workspace:
Version History
- Added main script mlx-stt.sh for audio transcription; removed previous Python script mlx-stt.py. - Updated usage instructions to use the new shell script instead of a Python command. - Expanded triggers in SKILL.md for easier activation. - Added version and author information to metadata. - The skill still performs local speech-to-text on Apple Silicon with MLX and open-source models.
What This Skill Does
Runs speech-to-text transcription locally on Apple Silicon using MLX and open-source models. No API key or external server needed. Default model is GLM-ASR-Nano-2512.
Runs entirely on-device with no usage costs or data leaving the machine, unlike cloud STT APIs.
When to use it
- Transcribing recorded meetings or interviews
- Converting voice memos to text
- Captioning locally stored video files
- Transcribing podcast episodes offline
- Turning audio notes into searchable text
Example Workflow
Here's how your AI assistant might use this skill in practice.
User asks: Transcribe my interview recording at ~/Downloads/interview.m4a
- 1Receives the audio file path from the user
- 2Runs mlx-stt.sh with the provided file path
- 3ffmpeg converts the audio format if needed
- 4MLX loads the GLM-ASR-Nano-2512 model (downloads on first run)
- 5Returns transcript printed to stdout
Full text transcript of the audio file
Requirements
Accounts, API keys, or tools you or your AI assistant may need to set up while using this skill.