Static project website and samples for MOSS-TTS-Nano, a 100M-parameter multilingual speech generation model with CPU inference and voice cloning.
Explore the website · Try the Hugging Face demo · Model downloads · Run the model locally
From the repository root, serve the static files with Python 3:
python -m http.server 4173Open localhost:4173. The page is in index.html, presentation styles are in styles.css, interactions are in script.js, and media files are in assets/.
This repository serves the project page. For inference, ONNX CPU deployment, voice cloning, and fine-tuning, follow the MOSS-TTS-Nano README.
Report website problems in this repository. Report model or inference problems in MOSS-TTS-Nano.