SOTAVerified

Speech Separation

The task of extracting all overlapping speech sources in a given mixed speech signal refers to the Speech Separation. Speech Separation is a special scenario of source separation problem, where the focus is only on the overlapping speech signal sources and other interferences such as music or noise signals are not the main concern of the study. A recent representative Github project can be referred to ClearerVoice-Studio.

Source: A Unified Framework for Speech Separation

Image credit: Speech Separation of A Target Speaker Based on Deep Neural Networks

Papers

Showing 101125 of 359 papers

TitleStatusHype
End-to-End Integration of Speech Separation and Voice Activity Detection for Low-Latency Diarization of Telephone Conversations0
Towards Real-Time Single-Channel Speech Separation in Noisy and Reverberant Environments0
Online Binaural Speech Separation of Moving Speakers With a Wavesplit Network0
Learning-based Robust Speaker Counting and Separation with the Aid of Spatial Coherence0
A Multi-Stage Triple-Path Method for Speech Separation in Noisy and Reverberant Environments0
Multi-Dimensional and Multi-Scale Modeling for Speech Separation Optimized by Discriminative Learning0
Scaling strategies for on-device low-complexity source separation with Conv-Tasnet0
MossFormer: Pushing the Performance Limit of Monaural Speech Separation using Gated Single-Head Transformer with Convolution-Augmented Joint Self-AttentionsCode1
Unifying Speech Enhancement and Separation with Gradient Modulation for End-to-End Noise-Robust Speech SeparationCode1
Deep AHS: A Deep Learning Approach to Acoustic Howling Suppression0
Short-Term Memory Convolutions0
Separate And Diffuse: Using a Pretrained Diffusion Model for Improving Source Separation0
Improving Target Speaker Extraction with Sparse LDA-transformed Speaker Embeddings0
Multi-resolution location-based training for multi-channel continuous speech separation0
An Audio-Visual Speech Separation Model Inspired by Cortico-Thalamo-Cortical CircuitsCode1
MIMO-DBnet: Multi-channel Input and Multiple Outputs DOA-aware Beamforming Network for Speech Separation0
Deep neural network techniques for monaural speech enhancement: state of the art analysis0
Deep Neural Mel-Subband Beamformer for In-car Speech Separation0
Self-Remixing: Unsupervised Speech Separation via Separation and Remixing0
Reverberation as Supervision for Speech Separation0
Handling Trade-Offs in Speech Separation with Sparsely-Gated Mixture of Experts0
An Adapter based Multi-label Pre-training for Speech Separation and Enhancement0
Speech separation with large-scale self-supervised learning0
Spatially Selective Deep Non-linear Filters for Speaker Extraction0
SCA: Streaming Cross-attention Alignment for Echo Cancellation0
Show:102550
← PrevPage 5 of 15Next →

Benchmark Results

#ModelMetricClaimedVerifiedStatus
1TF-Locoformer (L) + DMSI-SDRi25.1Unverified
2SepReformer-LSI-SDRi25.1Unverified
3TF-Locoformer (M) + DMSI-SDRi24.6Unverified
4TF-Locoformer (L)SI-SDRi24.2Unverified
5MossFormer2 (L)SI-SDRi24.1Unverified
6SepTDA (L=12)SI-SDRi24Unverified
7Separate And DiffuseSI-SDRi23.9Unverified
8TF-Locoformer (M)SI-SDRi23.6Unverified
9MossFormer (L) + DMSI-SDRi22.8Unverified
10TF-Locoformer (S) + DMSI-SDRi22.8Unverified
#ModelMetricClaimedVerifiedStatus
1TF-Locoformer (M)SI-SDRi18.5Unverified
2TF-Locoformer (S)SI-SDRi17.4Unverified
3SepReformer-L + DMSI-SDRi17.1Unverified
4MossFormer2SI-SDRi17Unverified
5MossFormer (L) + DMSI-SDRi16.3Unverified
6TD-Conformer (XL) + DMSI-SDRi14.6Unverified
7Improved Sudo rm -rf (U=36)SI-SDRi13.5Unverified
8TD-Conformer (L) + DMSI-SDRi13.4Unverified
9WavesplitSI-SDRi13.2Unverified
10DPTNET - SRSSNSI-SDRi12.3Unverified
#ModelMetricClaimedVerifiedStatus
1MossFormer2 (w speed perturb)SI-SDRi22.2Unverified
2TF-Locoformer (M)SI-SDRi22.1Unverified
3MossFormer2 (w/o DM)SI-SDRi21.7Unverified
4Separate And DiffuseSI-SDRi21.5Unverified
5WHYVSI-SDRi17.5Unverified
6TDANet LargeSI-SDRi17.4Unverified
7TDANetSI-SDRi16.9Unverified
8Conv-Tasnet (Libri1Mix speech enhancement pre-trained)SI-SDRi14.1Unverified
9Conv-Tasnet (Libri1Mix speech enhancement multi-task)SI-SDRi13.7Unverified
10Conv-TasnetSI-SDRi13.2Unverified
#ModelMetricClaimedVerifiedStatus
1SepTDASI-SDRi23.7Unverified
2MossFormer2SI-SDRi22.2Unverified
3MossFormer (L) + DMSI-SDRi21.2Unverified
4Separate And DiffuseSI-SDRi20.9Unverified
5MossFormer (M) + DMSI-SDRi20.8Unverified
6SepItSI-SDRi20.1Unverified
7SepFormerSI-SDRi19.5Unverified
8SandglassetSI-SDRi17.1Unverified
9Gated DualPathRNNSI-SDRi16.85Unverified
#ModelMetricClaimedVerifiedStatus
1IIANetSI-SNRi16.4Unverified
2TDFNet-largeSI-SNRi15.8Unverified
3TDFNet (MHSA + Shared)SI-SNRi15Unverified
4RTFS-Net-12SI-SNRi14.9Unverified
5RTFS-Net-6SI-SNRi14.6Unverified
6CTCNetSI-SNRi14.3Unverified
7RTFS-Net-4SI-SNRi14.1Unverified
8TDFNet-smallSI-SNRi13.6Unverified
#ModelMetricClaimedVerifiedStatus
1SepReformer-L + DMSI-SDRi18.4Unverified
2MossFormer2SI-SDRi18.1Unverified
3MossFormer (L) + DMSI-SDRi17.3Unverified
4TDANet LargeSI-SDRi15.2Unverified
5TDANetSI-SDRi14.8Unverified
6WHYVSI-SDRi12.96Unverified
#ModelMetricClaimedVerifiedStatus
1SepTDASI-SDRi21Unverified
2Hungarian PITSI-SDRi13.22Unverified
3Conditional TasNetSI-SDRi11.7Unverified
4TasTasSI-SDRi11.14Unverified
5Gated DualPathRNNSI-SDRi10.56Unverified
6Multi-Decoder DPRNNSI-SDRi5.9Unverified
#ModelMetricClaimedVerifiedStatus
1IIANetSI-SNRi18.3Unverified
2RTFS-Net-12SI-SNRi17.5Unverified
3CTCNetSI-SNRi17.4Unverified
4RTFS-Net-6SI-SNRi16.9Unverified
5RTFS-Net-4SI-SNRi15.5Unverified
#ModelMetricClaimedVerifiedStatus
1IIANetSI-SNRi14Unverified
2RTFS-Net-12SI-SNRi12.4Unverified
3CTCNetSI-SNRi11.9Unverified
4RTFS-Net-6SI-SNRi11.8Unverified
5RTFS-Net-4SI-SNRi11.5Unverified
#ModelMetricClaimedVerifiedStatus
1SepTDASI-SDRi22Unverified
2Gated DualPathRNNSI-SDRi12.88Unverified
3Conditional TasNetSI-SDRi12.5Unverified
4OR-PITSI-SDRi10.2Unverified
5Multi-Decoder DPRNNSI-SDRi9.3Unverified
#ModelMetricClaimedVerifiedStatus
1Separate And DiffuseSI-SDRi14.2Unverified
2SepItSI-SDRi13.7Unverified
3OCDSI-SDRi13.4Unverified
4Hungarian PITSI-SDRi12.72Unverified
#ModelMetricClaimedVerifiedStatus
1Separate And DiffuseSI-SDRi9Unverified
2SepItSI-SDRi8.2Unverified
3Hungarian PITSI-SDRi7.78Unverified
#ModelMetricClaimedVerifiedStatus
1SDR9.6Unverified
2Audio-Visual concat-refSDR8.05Unverified
#ModelMetricClaimedVerifiedStatus
1Separate And DiffuseSI-SDRi5.2Unverified
2Hungarian PITSI-SDRi4.26Unverified
#ModelMetricClaimedVerifiedStatus
1Conformer (base)0S5.6Unverified
2Conformer (large)0S5.4Unverified
#ModelMetricClaimedVerifiedStatus
1Hungarian PITSI-SDRi5.66Unverified
#ModelMetricClaimedVerifiedStatus
1Audio-Visual concat-refSDR10.55Unverified
#ModelMetricClaimedVerifiedStatus
1MossFormer2SI-SDRi20.5Unverified