List your company

Audio Captioning on ASFx (eval)

19.36SPIDErWhisper-Cards
0.97285.746410.5215.2936Feb 14, 2026

Evaluation results

MethodLinks
Whisper-CardsTraining data=Pixtral audiocards (descriptive caption)2026.0219.3653.4
Whisper-CardsTraining data=Pixtral audiocards (full card)2026.0218.6151.78
Audio Flamingo 3Mode=think2026.029.6142.61
GAMA2026.029.345.7
Whisper-BaselineTraining data=Pixtral captions2026.027.9849.78
Audio Flamingo 32026.025.2732.1
WavCaps HTSAT2026.023.9835.27
WavCaps CNN142026.02332.2
Audio Flamingo 22026.021.6830.08