Urban Sound Classification on UrbanSound8K (test)
88.8AccuracyMUKA
Evaluation Results
| Method | Links | |
|---|---|---|
| MUKATraining Approach=Training-free, Shots=16, Backbone=PENGI2026.02 | 88.8 | |
| Linear ProbingTraining Approach=Training-based, Shots=16, Backbone=PENGI2026.02 | 85.64 | |
| CLAP (Microsoft)Model Category=BM2025.11 | 82.74 | |
| WavCapsModel Category=BM2025.11 | 82.28 | |
| LAION-CLAPzero-shot=true2024.06 | 81.42 | |
| CLAP (LAION)Model Category=BM2025.11 | 80.91 | |
| PaLMTraining Approach=Training-based, Shots=16, Backbone=PENGI2026.02 | 80.77 | |
| CACARAComponents=WC/AC/C, f=0.12025.11 | 79.51 | |
| LanguageBindModel Category=MM2025.11 | 79.24 | |
| Treff-AdapterTraining Approach=Training-free, Shots=16, Backbone=PENGI2026.02 | 78.1 | |
| CACARAComponents=WC/AC/C, f=0.22025.11 | 77.83 | |
| CACARAComponents=WC/AC/C2025.11 | 77.4 | |
| CACARAComponents=AA/WC/AC/C2025.11 | 77.04 | |
| CoCoOpTraining Approach=Training-based, Shots=16, Backbone=PENGI2026.02 | 76.52 | |
| CoOpTraining Approach=Training-based, Shots=16, Backbone=PENGI2026.02 | 75.48 | |
| CACARAComponents=ASC/AA/WC/AC/C, f=0.12025.11 | 75.4 | |
| Pengizero-shot=true2024.06 | 75.29 | |
| CACARAComponents=ASC/AA/WC/AC/C, f=0.22025.11 | 74.99 | |
| M2D-CLAPmasking_ratio=0.7, zero-shot=true2024.06 | 72.4 | |
| CACARAComponents=ASC/AA/WC/AC/C2025.11 | 70.04 | |
| AudioCLIPzero-shot=false, note=Used test task data during training2024.06 | 68.78 | |
| VASTModel Category=MM2025.11 | 68 | |
| Zero-ShotTraining Approach=Training-free, Shots=0, Backbone=PENGI2026.02 | 53.49 | |
| ImageBindModel Category=MM2025.11 | 48.2 | |
| WavCapszero-shot=true2024.06 | 44 | |
| Wav2CLIPzero-shot=true2024.06 | 40.44 | |
| CLAP2023zero-shot=true2024.06 | 30 | |
| Proto-LCzero-shot=true2024.06 | 23.08 |