Speech Separation On Wham

Metrics

SI-SDRi

Results

Performance results of various models on this benchmark

		Paper Title
SepReformer-L + DM	18.4	Separate and Reconstruct: Asymmetric Encoder-Decoder for Speech Separation
MossFormer2	18.1	MossFormer2: Combining Transformer and RNN-Free Recurrent Network for Enhanced Time-Domain Monaural Speech Separation
MossFormer (L) + DM	17.3	MossFormer: Pushing the Performance Limit of Monaural Speech Separation using Gated Single-Head Transformer with Convolution-Augmented Joint Self-Attentions
TDANet Large	15.2	An efficient encoder-decoder architecture with top-down attention for speech separation
TDANet	14.8	An efficient encoder-decoder architecture with top-down attention for speech separation
WHYV	12.964	An alternative Approach in Voice Extraction

0 of 6 row(s) selected.