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

UltraEval-Audio

Benchmarks

Task NameDataset NameSOTA ResultTrend
Spoken Question AnsweringUltraEval-Audio WebQ
S2T Score (Task Correctness)55.95
9
Spoken Question AnsweringUltraEval-Audio LLaMA-QA
S2T Score83
9
Spoken Question AnsweringUltraEval-Audio S2S
AlpacaEval Score0.7338
9
Showing 3 of 3 rows