qwen3-tts-apple-silicon

Runs the Qwen3 text-to-speech model offline on Apple Silicon Macs. It supports voice cloning and custom voices without cloud services.

Share on XNo license declared

Overview

This project runs the Qwen3-TTS text-to-speech model locally on Apple Silicon Macs (M1 to M4) using Apple's MLX framework. It works fully offline, with no cloud service or API keys. It supports cloning a voice from a short sample, designing new voices from a text description, and using built-in voices with emotion and speed control. The README says MLX models use less RAM than standard PyTorch ones.

Key features

  • Voice cloning from a 5-second audio sample
  • Voice design from a text description
  • 9 built-in voices with emotion and speed control
  • Runs offline on M-series Macs with MLX
  • No cloud service or API keys

Best for

Mac users who want private, offline text-to-speech and voice cloning. It targets Apple Silicon only.

Upstream
kapi2800/qwen3-tts-apple-silicon
Fork on GitHub
Guo-astro/qwen3-tts-apple-silicon
Upstream stars
570
Category
Documents, media and content
Language
Python
License
No license declaredWithout a license, the author keeps all rights. Ask the upstream owner before reusing the code.
Forked
2026-02-18
Sync status
In syncLast synced 2026-09-29