MegaTTS3 Voice Cloning Node for ComfyUI

High-quality voice cloning, supports Chinese and English, and can perform cross-lingual cloning.

📣 Updates

[2025-04-28]⚒️: Add a preview voice node, preview the voice first, and then clone if satisfied. Thank you @chenpipi0807 for your idea 😍. You can create more folders by categorizing them under the speakers folder.

[2025-04-06]⚒️: Released v1.0.0.

Installation

cd ComfyUI/custom_nodes
git clone https://github.com/billwuhao/ComfyUI_MegaTTS3.git
cd ComfyUI_MegaTTS3
pip install -r requirements.txt

# python_embeded
./python_embeded/python.exe -m pip install -r requirements.txt

Model Download

Models and voices need to be manually downloaded and placed in the ComfyUI\models\TTS directory:

MegaTTS3 Download the entire folder and place it in the TTS folder.

Create a new speakers folder inside the MegaTTS3 folder. Download all .wav and .npy files from Google drive and place them in the speakers folder.

Acknowledgements

MegaTTS3