ResearchDatasetsDVSpeakerDVSpeakerFollowBenchmarksTaskDatasetSOTA methodSortMost resultsRecently updatedMost papersApplyTaskDatasetSOTA resultTrendResultsVisual Speech RecognitionDVSpeaker (Cross-scene)71.68SI (45°)21Visual Speech RecognitionDVSpeaker Matched-scene100SI Accuracy (0°)21