{"skill":{"slug":"mlx-stt","displayName":"MLX STT","summary":"Speech-To-Text with MLX (Apple Silicon) and opensource models (default GLM-ASR-Nano-2512) locally.","description":"---\nname: mlx-stt\ndescription: Speech-To-Text with MLX (Apple Silicon) and opensource models (default GLM-ASR-Nano-2512) locally.\nversion: 1.0.7\nauthor: guoqiao\nmetadata: {\"openclaw\":{\"always\":true,\"emoji\":\"🦞\",\"homepage\":\"https://github.com/guoqiao/skills/blob/main/mlx-stt/mlx-stt/SKILL.md\",\"os\":[\"darwin\"],\"requires\":{\"bins\":[\"brew\"]}}}\ntriggers:\n- \"/mlx-stt <audio>\"\n- \"STT ...\"\n- \"ASR ...\"\n- \"Transcribe ...\"\n- \"Convert audio to text ...\"\n---\n\n# MLX STT\n\nSpeech-To-Text/ASR/Transcribe with MLX (Apple Silicon) and opensource models (default GLM-ASR-Nano-2512) locally.\n\nFree and Accurate. No api key required. No server required.\n\n## Requirements\n\n- `mlx`: macOS with Apple Silicon\n- `brew`: used to install deps if not available\n\n## Installation\n\n```bash\nbash ${baseDir}/install.sh\n```\nThis script will use `brew` to install these cli tools if not available:\n- `ffmpeg`: convert audio format when needed\n- `uv`: install python package and run python script\n- `mlx_audio`: do the real job\n\n## Usage\n\nTo transcribe an audio file, run this script:\n\n```bash\nbash  ${baseDir}/mlx-stt.sh <audio_file_path>\n```\n\n- First run could be a little slow, since it will need to download model.\n- The transcript result will be printed to stdout.\n","tags":{"Apple Silicon":"1.0.7","Mac mini":"1.0.7","MacBook":"1.0.7","asr":"1.0.7","audio":"1.0.7","glm":"1.0.7","glm-asr":"1.0.7","glm-asr-nano-2512":"1.0.7","glm-asr-nano-2512-8bit":"1.0.7","latest":"1.0.7","macOS":"1.0.7","mlx":"1.0.7","mlx-audio":"1.0.7","speech-to-text":"1.0.7","stt":"1.0.7","darwin":"1.0.1"},"stats":{"comments":1,"downloads":4148,"installsAllTime":17,"installsCurrent":17,"stars":1,"versions":8},"createdAt":1769921433962,"updatedAt":1778485932126},"latestVersion":{"version":"1.0.7","createdAt":1770686109778,"changelog":"- Added main script mlx-stt.sh for audio transcription; removed previous Python script mlx-stt.py.\n- Updated usage instructions to use the new shell script instead of a Python command.\n- Expanded triggers in SKILL.md for easier activation.\n- Added version and author information to metadata.\n- The skill still performs local speech-to-text on Apple Silicon with MLX and open-source models.","license":null},"metadata":{"setup":[],"os":["darwin"],"systems":null},"owner":{"handle":"guoqiao","userId":"s17ffveb731pj7ry6ezhyn9hts88598a","displayName":"guoqiao","image":"https://avatars.githubusercontent.com/u/1306111?v=4"},"moderation":null}