Our new X account is live! Follow @wizwand_team for updates
WorkDL logo mark

Verse-Bench

Benchmarks

Task NameDataset NameSOTA ResultTrend
Text-to-Audio-Video GenerationVerse-Bench
MS0.58
16
Audio-Visual GenerationVerse-Bench (All subsets)
IS (Score)4.269
7
Audio-Visual GenerationVerse-Bench multi-speaker
cpCER14.9
6
Audio-Visual GenerationVerse-Bench (set3)
DNSMOS3.797
6
Talking Head GenerationVerse-Bench
LSE-C1.62
5
Showing 5 of 5 rows