wav2vec2-base-100k-gtzan-music-genres
| Property | Value |
|---|---|
| Author | m3hrdadfi |
| Framework | PyTorch |
| Task | Audio Classification |
| Overall Accuracy | 77.5% |
What is wav2vec2-base-100k-gtzan-music-genres?
This is a specialized audio classification model built on the Wav2Vec 2.0 architecture, designed specifically for music genre classification. The model can classify music into 10 distinct genres including blues, classical, country, disco, hiphop, jazz, metal, pop, reggae, and rock.
Implementation Details
The model utilizes the Wav2Vec 2.0 feature extractor and requires PyTorch for implementation. It processes audio inputs and provides probability scores for each music genre category. The implementation includes resampling capabilities to ensure compatibility with various input formats.
- Built on Wav2Vec 2.0 architecture
- Supports 10 music genres
- Includes automatic audio resampling
- Provides probability scores for each genre
Core Capabilities
- Highest accuracy for Jazz classification (100% precision)
- Strong performance in Classical music (90.5% F1-score)
- Effective genre disambiguation with probability scoring
- Real-time audio processing and classification
Frequently Asked Questions
Q: What makes this model unique?
The model combines the powerful Wav2Vec 2.0 architecture with specialized training for music genre classification, achieving particularly high accuracy in jazz and classical genres while maintaining robust performance across all categories.
Q: What are the recommended use cases?
This model is ideal for music library organization, automated playlist creation, music recommendation systems, and research applications requiring genre classification. It's particularly effective for classical and jazz music identification.
