This repository contains the code and resources for a Music Genre Classification project based on Multi-Class Support Vector Machines (SVM). The goal of the project is to build a robust model capable of classifying music tracks into various genres using datasets like the GTZAN dataset for training and validation.
The GTZAN dataset is utilized for training and validating the SVM model. The dataset includes audio files from different genres, providing a diverse set of samples for effective model training.
-
SVM Model: Implementation of a Multi-Class Support Vector Machines model for music genre classification.
-
Data Preprocessing: Extraction of relevant audio features, such as Mel-frequency cepstral coefficients (MFCCs), spectral contrast, and chroma features.
-
Evaluation: Performance evaluation metrics, including accuracy, precision, recall, and F1-score, are provided to assess the model's effectiveness.
Detailed results, including confusion matrices and classification reports, are available in the 'results' folder.
- Exploration of advanced feature extraction techniques.
- Fine-tuning hyperparameters for optimal model performance.
- Integration of deep learning models for comparison.
Feel free to contribute, report issues, or suggest improvements!