| emotion-recognition-on-emomusic | M2D-CLAP | #1 | EmoA: 77.4EmoV: 61.9 |
| emotion-recognition-on-emomusic | M2D2 | #2 | EmoA: 76.7EmoV: 59.3 |
| emotion-recognition-on-emomusic | M2D | #3 | EmoA: 76.1EmoV: 59.4 |
| instrument-recognition-on-nsynth | M2D-CLAP | #1 | Accuracy: 80.6 |
| instrument-recognition-on-nsynth | M2D2 AS+ | #2 | Accuracy: 79.7 |
| instrument-recognition-on-nsynth | M2D AS | #3 | Accuracy: 78.7 |
| music-auto-tagging-on-magnatagatune | M2D2 AS+ | #1 | PR-AUC: 41.6ROC AUC: 91.8 |
| audio-classification-on-esc-50 | M2D2 AS+ | #3 | Top-1 Accuracy: 98.5PRE-TRAINING DATASET: AudioSet,WavCaps… |
| audio-classification-on-audioset | M2D2 | #14 | Test mAP: 0.490 |