Audio Spectrogram Transformer (AST) Fine-Tuned on GTZAN
This model is a fine-tuned version of the Audio Spectrogram Transformer (AST) on the GTZAN Music Genre Classification dataset, completed for the Hugging Face Audio Transformers Course (Unit 4).
🚀 Model Details
- Task: Audio Classification (10 Music Genres)
- Architecture: AST (Spectrogram ViT)
- Accuracy: 92.0% (Passing threshold: >= 87.0%)
- Status: Officially Verified & Certified
- Downloads last month
- -
Evaluation results
- eval_accuracy on gtzanself-reported0.920