ResearchTasksAudio-driven half-body human video generationAudio-driven half-body human video generationFollowBenchmarksDatasetSOTA methodSortMost resultsRecently updatedMost papersApplyDatasetSOTA methodMetricTrendResultsLast updatedEMTD 1.0 (evaluation set)EchoMimicV249.33FID14Feb 26, 2026