An Audio-Visual Speech Separation Model Inspired by Cortico-Thalamo-Cortical Circuits

Open paper

Benchmark	Model	Rank	Results
speech-separation-on-lrs3	CTCNet	#3	SI-SNRi: 17.4
speech-separation-on-voxceleb2	CTCNet	#3	SI-SNRi: 11.9
speech-separation-on-lrs2	CTCNet	#6	SI-SNRi: 14.3