A high-performance, local speech-to-text solution optimized for Apple Silicon using MLX and open-source models.
Install & Download
1. ClawHub CLI
The fastest way to install a skill directly from the registry.
npx clawhub@latest install mlx-stt
2. Manual Installation
Copy the skill folder to one of these locations
~/.openclaw/skills/ <project>/skills/ Priority: Workspace > Local > Bundled
3. Prompt Installation
Copy this prompt to OpenClaw to install it automatically.
Help me install mlx-stt using Clawhub. If Clawhub is not installed, install it first (npm i -g clawhub).
Prefer to download?
Get the raw skill files in a ZIP archive.
What is mlx-stt?
The mlx-stt skill is a powerful local Automatic Speech Recognition (ASR) tool designed specifically for macOS users with Apple Silicon. By leveraging the MLX framework and open-source models like GLM-ASR-Nano-2512, this skill provides fast and accurate transcription without the need for external API keys or cloud servers. It is an essential addition to your collection of Openclaw Skills for maintaining data privacy while processing audio content and ensuring high-speed local execution.
mlx-stt Use Cases
- Transcribing voice memos or audio recordings directly on your Mac without internet access.
- Converting recorded meetings or interviews into text for immediate documentation.
- Automating transcription workflows within your local environment using Openclaw Skills.
- Processing sensitive audio data locally to ensure maximum privacy and security.
How mlx-stt Works
- The skill is triggered via a specific command or by providing an audio file path.
- It verifies the local environment for dependencies like ffmpeg and the MLX library.
- Upon the first execution, it automatically downloads the required open-source ASR models.
- Audio processing is performed using the MLX framework, which is hardware-accelerated for Apple Silicon.
- The final transcribed text is output directly to the console for further use or logging.
mlx-stt Setup
To get started with this skill on your Apple Silicon Mac, ensure you have Homebrew installed, then run the installation script:
bash ${baseDir}/install.sh
This script handles the setup of ffmpeg, uv, and mlx_audio automatically. To transcribe a file, use:
bash ${baseDir}/mlx-stt.sh <audio_file_path>
mlx-stt Data Schema & Taxonomy
The mlx-stt skill handles audio inputs and model data locally. Here is how the technical components are organized:
| Component | Type | Description |
|---|---|---|
| Input | File Path | Local path to the audio file (converted via ffmpeg if necessary). |
| Model | MLX Model | Default: GLM-ASR-Nano-2512, stored in local cache. |
| Output | String | Plain text transcription sent to stdout. |
| Tools | CLI | Utilizes ffmpeg for conversion and uv for Python environment management. |
mlx-stt Advanced Features
- Hardware-accelerated inference specifically tuned for M-series Apple chips.
- Support for multiple transcription triggers including /mlx-stt, ASR, and Transcribe.
- Completely offline operation, requiring no external API calls after the initial model download.
- Seamless integration into broader automation pipelines powered by Openclaw Skills.
SKILL.md
Loading
Related Openclaw Skills
MLX Audio Server
A local, 24/7 OpenAI-compatible API server for high-performance STT and TTS powered by the MLX framework on macOS.
HackerNews Extract
A utility that extracts a HackerNews post, including the linked article and all threaded comments, into a single, well-formatted Markdown document.
Media Downloader
A smart media retrieval skill that downloads video and audio from various platforms directly into organized local folders.
Zhihu Hot List Monitor
A specialized monitoring tool designed to fetch, filter, and analyze the most popular trending topics and questions on Zhihu.
MLX TTS
A high-performance local text-to-speech engine for macOS that leverages Apple Silicon and MLX for fast, private audio synthesis.
URL to PDF
A high-performance tool for converting URLs into mobile-optimized PDF documents for easy reading.
METADATA
Featured*
