OpenCodePapers

speech-separation-on-libri2mix

Speech Separation
Dataset Link
Results over time
Click legend items to toggle metrics. Hover points for model names.
Leaderboard
Click a sortable column header to sort.
PaperCodeSI-SDRiSDRiNumber of parameters (M)SDRModelNameReleaseDate
MossFormer2: Combining Transformer and RNN-Free Recurrent Network for Enhanced Time-Domain Monaural Speech Separation✓ Link22.2MossFormer2 (w speed perturb)2023-12-19
TF-Locoformer: Transformer with Local Modeling by Convolution for Speech Separation and Enhancement✓ Link22.122.215TF-Locoformer (M)2024-08-06
MossFormer2: Combining Transformer and RNN-Free Recurrent Network for Enhanced Time-Domain Monaural Speech Separation✓ Link21.7MossFormer2 (w/o DM)2023-12-19
Separate And Diffuse: Using a Pretrained Diffusion Model for Improving Source Separation21.5Separate And Diffuse2023-01-25
Wanna hear your voice? A sample is all we need!17.517.2458WHYV2024-10-01
An efficient encoder-decoder architecture with top-down attention for speech separation✓ Link17.4TDANet Large2022-09-30
An efficient encoder-decoder architecture with top-down attention for speech separation✓ Link16.9TDANet2022-09-30
Stabilizing Label Assignment for Speech Separation by Self-supervised Pre-training✓ Link14.114.6Conv-Tasnet (Libri1Mix speech enhancement pre-trained)2020-10-29
Stabilizing Label Assignment for Speech Separation by Self-supervised Pre-training✓ Link13.714.1Conv-Tasnet (Libri1Mix speech enhancement multi-task)2020-10-29
Stabilizing Label Assignment for Speech Separation by Self-supervised Pre-training✓ Link13.213.6Conv-Tasnet2020-10-29
Check out how to contribute new results.