Follow
AV2Wav: Diffusion-Based Re-synthesis from Continuous Self-supervised Features for Audio-Visual Speech Enhancement. (arXiv:2309.08030v2 [eess.AS] UPDATED)
AV2Wav: Diffusion-Based Re-synthesis from Continuous Self-supervised Features for Audio-Visual Speech Enhancement. (arXiv:2309.08030v2 [eess.AS] UPDATED)