Audio Captioning on Clotho (eval)
31.88SPIDErAudio Flamingo 3
Evaluation Results
| Method | Links | ||
|---|---|---|---|
| Audio Flamingo 32026.02 | 31.88 | 52.11 | |
| WavCaps CNN142026.02 | 31 | — | |
| WavCaps HTSAT2026.02 | 29.7 | — | |
| Whisper-CardsTraining data=Pixtral audiocards (descriptive caption)2026.02 | 22.18 | 48.48 | |
| Whisper-CardsTraining data=Pixtral audiocards (full card)2026.02 | 22.07 | 48.67 | |
| Whisper-BaselineTraining data=Pixtral captions2026.02 | 21.05 | 47.61 | |
| Audio Flamingo 3Mode=think2026.02 | 13.22 | 50.19 | |
| Audio Flamingo 22026.02 | 7.25 | 33.71 | |
| GAMA2026.02 | 5.34 | 43.21 |