ResearchTasksAudio-Visual Speech ExtractionAudio-Visual Speech ExtractionFollowBenchmarksDatasetSOTA methodSortMost resultsRecently updatedMost papersApplyDatasetSOTA methodMetricTrendResultsLast updatedVoxCeleb-SimIsoNet-CL19.31SI-SDR (dB)3May 15, 2026WSJ0-mixGu et al.12SI-SDR (dB)1May 15, 2026FAIR-PlayVisualVoice11.8SI-SDR (dB)1May 15, 2026AVSpeechLooking to Listen10.2SI-SDR (dB)1May 15, 2026