mlx-tts

Generate speech audio from text locally on macOS Apple Silicon using MLX models.

Updated Feb 1, 2026
One-click install
npx skills add https://github.com/guoqiao/skills --skill mlx-tts
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: mlx-tts
Source: https://github.com/guoqiao/skills/tree/main/mlx-tts/mlx-tts
Command: npx skills add https://github.com/guoqiao/skills --skill mlx-tts

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill requires mlx-audio, ffmpeg, and includes scripts (resource) components.

What problem does it solve?

This Skill provides a fast, free, and private way to convert text into speech directly on your local machine, without relying on external APIs or servers.

Core Features & Use Cases

  • Local TTS: Utilizes MLX and open-source models (default QWen3-TTS) for text-to-speech.
  • Apple Silicon Optimized: Designed to run efficiently on macOS with Apple Silicon.
  • No API Keys/Servers: Operates entirely offline, ensuring privacy and reducing costs.
  • Use Case: Quickly generate voice replies for messages or create audio versions of important text snippets without sending data to the cloud.

Quick Start

Convert the text "Hello, world!" into an audio file using the mlx-tts skill.

Frequently Asked Questions about mlx-tts

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I run local text-to-speech synthesis on Apple Silicon without an internet connection?

Local text-to-speech synthesis on Apple Silicon runs entirely offline using MLX and open-source models. This on-device approach generates audio privately without relying on external APIs or servers.

Can I use MLX for speech synthesis on macOS without paying for API keys?

Yes, you can use MLX for speech synthesis on macOS without API keys. It operates entirely offline using open-source models, reducing costs while ensuring data privacy during audio generation.

Do I need ffmpeg installed to convert text to speech locally using mlx?

Yes, ffmpeg is required alongside mlx-audio to perform local text-to-speech synthesis. These dependencies handle model execution and audio format conversion on your macOS machine.

What is the best way to generate private voice replies from text on macOS?

The best way to generate private voice replies from text on macOS is using local speech synthesis with MLX. It processes everything on-device, ensuring privacy and quick audio generation.

Does local speech synthesis work offline with the Qwen3-TTS model?

Yes, local speech synthesis works offline with the Qwen3-TTS model. By utilizing MLX on Apple Silicon, the system executes on-device audio generation without needing cloud connectivity.

What are the limitations of running local TTS on Apple Silicon?

Limitations of running local TTS on Apple Silicon include being restricted to macOS hardware. It requires specific dependencies like mlx-audio and ffmpeg, and is tailored for Apple Silicon efficiency.