Audio-Visual Speech Enhancement Method Conditioned in the Lip Motion and Speaker-Discriminative Embeddings.

Koichiro Ito, Masaaki Yamamoto, Kenji Nagamatsu

Browse the full ICASSP paper archive.