Music Genre Recognition on GTZAN
82.1Top-1 AccuracyCLIP ViT-B/32
Evaluation Results
| Method | Links | ||
|---|---|---|---|
| CLIP ViT-B/32Backbone=ViT-B/322026.06 | 82.1 | — | |
| DinoV3 ViT-SBackbone=ViT-S2026.06 | 70.2 | — | |
| PlanFormerProtocol=Linear probing (5-fold CV), Classes=10, Type=Semantic2026.06 | 64.1 | 89.4 | |
| Expert features (MFCC) + linear SVMType=Expert Features2026.06 | 62.5 | — | |
| Random Init.Classes=10, Type=Semantic2026.06 | 47.1 | — |