Skip to content

Conversation

@rafipatel
Copy link

@rafipatel rafipatel commented Nov 13, 2025

Description:
This PR introduces a streamlined setup and deployment workflow for MuseTalk V15, including:

  • V15 Colab notebook: For fast cloud experimentation and demo.
  • FastAPI serving: Enables RESTful inference via a Python API (usable in both local and production setups).
  • Automated model downloads (download_models.py): Robust pure-Python script to fetch all required weights from Hugging Face, Google Drive, and direct URLs, with auto package installation.
  • Dockerfile: Provides working containerization, verified on Mac M3 hardware.

Why was this needed?
The previous download_weights.sh script frequently failed due to issues with huggingface-cli, including authentication errors and version mismatches across environments.
This PR replaces it with a portable Python script that is more reliable, does not require manual CLI tooling, and works "out of the box" for both local development and Docker-based deployments.

@rafipatel rafipatel changed the title Added V15 Colab notebook, FastAPI Serving, model downloads .py script, and Dockerfile (working in MAC M3) Added V15 Colab notebook, FastAPI Serving, model downloads .py script, and Dockerfile (working in MAC M3), readme updates Nov 13, 2025
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.

1 participant