🎵 Audio · Free

Microsoft VibeVoice TTS

Run Microsoft's VibeVoice text-to-speech locally on Windows using either Gradio or ComfyUI, with custom voice-cloning support. The original Microsoft repository was taken down, so this guide uses a maintained backup fork.

✓ 100% free — download the workflow and run it locally.
⬇ Download workflow (.json) ▶ Watch tutorial

Requirements

Model downloads

Custom nodes

Setup steps

  1. Gradio: clone the fork — git clone https://github.com/TheLocalLab/VibeVoice-Fork.git
  2. Gradio: install it — pip install -e .
  3. Install PyTorch (CUDA 12.8) — pip install torch torchvision torchaudio --index-url https://download.pytorch.org/whl/cu128
  4. Download a Flash Attention wheel from github.com/kingbri1/flash-attention/releases and install it — pip install <wheel_filename>
  5. Install Triton — pip install -U "triton-windows<3.5"
  6. ComfyUI: clone both custom nodes into ComfyUI/custom_nodes/ and install their requirements.
  7. Download model weights to ComfyUI/models/tts/VibeVoice/, launch ComfyUI, and load the example workflow below.

Notes & tips

Want the 1-click version?

Skip the manual model downloads and node installs — get a ready-to-run installer, or unlock all 75+ with Local Lab Pro.