search

Sindhu-Hegde / pseudo-visual-speech-denoising

Official code for the paper "Visual Speech Enhancement Without A Real Visual Stream" published at WACV 2021
MIT License
103 stars 24 forks source link

Could you tell me what would be the function of tmp.wav in the inference.py code...thank you #6

Closed SGT103 closed 2 years ago

Sindhu-Hegde commented 3 years ago

The "tmp.wav" file (line: https://github.com/Sindhu-Hegde/pseudo-visual-speech-denoising/blob/main/inference.py#L17) extracts the audio from the input noisy (audio/video) file. This is to ensure the wav file is extracted from any other input format (example mp4).