HyperAIHyperAI
HomeConsoleDocsNewsPapersTutorialsDatasetsWikiSOTALLM ModelsGPU LeaderboardEvents
Search
About
HyperAIHyperAI

Command Palette

Search for a command to run...

Music Source Separation On Musdb18 Hq

Metrics

SDR (avg)
SDR (bass)
SDR (drums)
SDR (others)
SDR (vocals)

Results

Performance results of various models on this benchmark

Paper TitleRepository
BS-RoFormer (L=12, OA)11.9913.3212.919.0112.72--
BS-RoFormer (L=6, OA)9.8011.319.497.7310.66--
SCNet-large9.699.4910.987.4410.86SCNet: Sparse Compression Network for Music Source Separation
Sparse HT Demucs (fine tuned)9.2010.4710.836.419.37Hybrid Transformers for Music Source Separation
SCNet9.008.8210.516.769.89SCNet: Sparse Compression Network for Music Source Separation
Hybrid Transformer Demucs (f.t.)9.0010.3910.086.329.20Hybrid Transformers for Music Source Separation
Band-Split RNN (semi-sup.)8.978.1610.157.0810.47Music Source Separation with Band-split RNN
TFC-TDF-UNet (v3)8.348.458.446.869.59Sound Demixing Challenge 2023 Music Demixing Track Technical Report: TFC-TDF-UNet v3
Band-Split RNN8.247.229.016.7010.01Music Source Separation with Band-split RNN
Dual-Path TFC-TDF UNet (DTTNet)8.157.557.827.0210.21Music Source Separation Based on a Lightweight Deep Learning Framework (DTTNET: DUAL-PATH TFC-TDF UNET)
Hybrid Demucs7.688.768.245.598.13Hybrid Spectrogram and Waveform Source Separation
KUIELab-MDX-Net7.477.837.205.908.97KUIELab-MDX-Net: A Two-Stream Neural Network for Music Demixing
CWS-PResUNet6.775.936.385.848.92CWS-PResUNet: Music Source Separation with Channel-wise Subband Phase-aware ResUNet
Unmix4.1884.0734.9252.6955.06Transfer Learning with Jukebox for Music Source Separation
0 of 14 row(s) selected.
Music Source Separation On Musdb18 Hq | SOTA | HyperAI