Repository for the paper "Disentanglement Learning for Variational Autoencoders Applied to Audio-Visual Speech Enhancement".
Link to the paper: https://arxiv.org/abs/2105.08970
Link to visual-only VAD: https://github.com/sp-uhh/audio-visual-vad
Audio and spectrogram examples are provided here: https://uhh.de/inf-sp-disentangled2021