Share your thoughts, 1 month free Claude Pro on usSee more
WorkDL logo mark

MultiDialog

Benchmarks

Task NameDataset NameSOTA ResultTrend
Spoken DialogueMultiDialog 1.0 (test)
PPL930.401
8
Audio-visual generationMultiDialog (test)
SIM0.624
4
Fine-grained Score AccuracyMultiDialog
Exact Accuracy65.62
1
Binary classification (Human vs Machine speech)MultiDialog (Human-Human) OOD (test)
Accuracy95.31
1
Showing 4 of 4 rows