Audio-to-Text Retrieval on Clotho 1K 1.0 (test)
27.1R@1ONE-PEACE
Evaluation Results
| Method | Links | |||
|---|---|---|---|---|
| ONE-PEACE# (audio, text)=2.4M, Training Regime=internet-scale2023.12 | 27.1 | 52.3 | 65.4 | |
| LAION-CLAP# (audio, text)=700K, Training Regime=internet-scale2023.12 | 24.2 | 51.1 | 66.9 | |
| FuseMix(W&H,B)# (audio, text)=65K, Training Regime=low-data regime2023.12 | 21.5 | 44.7 | 57.4 | |
| CLAP-HTSAT# (audio, text)=70K, Training Regime=low-data regime2023.12 | 20 | 44.9 | 58.7 | |
| FuseMix(W&H,B)# (audio, text)=50K/15K, Training Regime=low-data regime2023.12 | 19.7 | 42.9 | 56.5 | |
| ML-ACT# (audio, text)=50K/15K, Training Regime=low-data regime2023.12 | 16.2 | 37.5 | 50.2 | |
| LAION-CLAP# (audio, text)=65K, Training Regime=low-data regime2023.12 | 15.7 | 36.9 | 51.3 | |
| MMT# (audio, text)=50K/15K, Training Regime=low-data regime2023.12 | 6.3 | 22.8 | 33.3 |