OpenCodePapers

speech-separation-on-wsj0-3mix

Speech Separation
Results over time
Click legend items to toggle metrics. Hover points for model names.
Leaderboard
Click a sortable column header to sort.
PaperCodeSI-SDRiModelNameReleaseDate
Boosting Unknown-number Speaker Separation with Transformer Decoder-based Attractor23.7SepTDA2024-01-23
MossFormer2: Combining Transformer and RNN-Free Recurrent Network for Enhanced Time-Domain Monaural Speech Separation✓ Link22.2MossFormer22023-12-19
MossFormer: Pushing the Performance Limit of Monaural Speech Separation using Gated Single-Head Transformer with Convolution-Augmented Joint Self-Attentions✓ Link21.2MossFormer (L) + DM2023-02-23
Separate And Diffuse: Using a Pretrained Diffusion Model for Improving Source Separation20.9Separate And Diffuse2023-01-25
MossFormer: Pushing the Performance Limit of Monaural Speech Separation using Gated Single-Head Transformer with Convolution-Augmented Joint Self-Attentions✓ Link20.8MossFormer (M) + DM2023-02-23
SepIt: Approaching a Single Channel Speech Separation Bound20.1SepIt2022-05-24
Attention is All You Need in Speech Separation✓ Link19.5SepFormer2020-10-25
Sandglasset: A Light Multi-Granularity Self-attentive Network For Time-Domain Speech Separation✓ Link17.1Sandglasset2021-03-01
Voice Separation with an Unknown Number of Multiple Speakers✓ Link16.85Gated DualPathRNN2020-02-29
Check out how to contribute new results.