Skip to content

Conversation

@chendaile
Copy link

  • Title
    Update requirements for Python 3.10 and CUDA 12.1 support
  • Summary
    • refresh dependency pins in requirements.txt for Python 3.10 and CUDA 12.1 torch wheels (requirements.txt:1)
    • replace deprecated librosa.output.write_wav with soundfile.write for audio export compatibility (audio.py:7, audio.py:18)
    • enhance checkpoint loading to handle TorchScript .pt files and state-dict .pth weights, plus directory
      inputs (inference.py:160, inference.py:164, inference.py:175)
  • Testing
    • python inference.py --checkpoint_path checkpoints/Wav2Lip-SD-NOGAN.pt --audio your/wav/path --face your/png/path

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