ResearchDatasetsDnRFollowBenchmarksTask NameDataset NameSOTA MethodSortMost resultsRecently updatedMost papersApplyTask NameDataset NameSOTA ResultTrendResultsCinematic Audio Source SeparationDnR (test)16.3Speech SISDR10Universal Sound Separationdnr v2 (test)6.8Music SI-SDR8Speech Target ExtractionDNR nonverbal13.01SDR (dB)7Source Separationdnr v2 (test)3.34Overall MOS-LQS Score2