跳到正文
TakoAPI
返回技能列表

mlx-audio-server

由 guoqiao 提供

Media & Streaming
2.2k下载量由 clawskills.sh 提供
0浏览量
0 个赞登录后即可为这个技能点赞
—暂无评分
你的评分

简介

A fast, accurate, and fully local OpenAI-compatible API.

描述

该技能的功能

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.

何时使用

  • 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

示例工作流

输入

User asks: transcribe this interview recording at ./interview.mp4

智能体
  1. 1

    Calls run_stt.sh with the video path

  2. 2

    ffmpeg converts the file to wav internally

  3. 3

    Server sends audio to local mlx-community/glm-asr-nano-2512-8bit model

  4. 4

    Transcript text is returned to stdout

输出

Plain text transcript of the interview

前置要求

  • Apple Silicon Mac (M1 or later)
  • Homebrew installed

安装

clawhub install guoqiao/mlx-audio-server

运行 CLI 命令,或把这段提示词粘贴到 Claude Code、Cursor、Codex 或你在用的任意编程智能体里。

链接