jotihicks/African-Music-Genre-Classifier-

An end-to-end pipeline for African music genre classification. Includes an automated agent for mass dataset acquisition and a Jupyter Notebook for model training and analysis. # African-Music-Genre-Classifier(AFMGC)- An end-to-end pipeline for African music genre classification. Includes an automated agent for mass dataset acquisition and a Jupyter Notebook for model training and analysis. ## 📖 Overview This repository contains the codebase for a deep learning project aimed at classifying distinct African music genres (e.g., Afrobeat, Highlife, Amapiano, Soukous, Chimurenga). The project consists of two main modules: **Classification Model:** A Jupyter Notebook containing the preprocessing logic, feature extraction (MFCCs/Spectrograms), and model training pipeline. **🎯 Target Genres** The model will be trained to distinguish between the following classes: Afrobeat Afrobeats Amapiano Highlife Makosa Kwaito Soukous Juju Fuji Congolese Rumba ## Features - Downloads audio from YouTube using `yt-dlp` - Converts to WAV format (16-bit PCM, configurable sample rate) - Organizes by genre in separate folders - Logs metadata (filename, genre, URL, title, duration) to CSV - Robust error handling - continues on failures - Configurable number of results per query ## Requirements - Python 3.7+ - ffmpeg (must be installed and in PATH) - yt-dlp - pandas ## Installation 1. Install ffmpeg: - Windows: Download from ffmpeg.org - Linux: `sudo apt install ffmpeg` - Mac: `brew install ffmpeg` 2. Install Python dependencies: ```bash pip install -r requirements.txt ``` ## Usage 1. Edit `search_queries.json` with your genres/search terms: ```json [ "Afrobeats Nigeria", "Highlife Ghana", "Fuji music Nigeria" ] ``` 2. Run the downloader: ```bash python download_agent.py ``` 3. Configure settings in the script: - `MAX_RESULTS_PER_QUERY`: Number of tracks per genre (default: 50) - `SAMPLE_RATE`: Audio sample rate - 22050 or 44100 Hz (default: 22050) ## Output Structure ``` african_music_dataset/ ├── Afrobeats_Nigeria/ │ ├── dQw4w9WgXcQ.wav │ └── ... ├── Highlife_Ghana/ │ └── ... ├── metadata.csv └── error_log.txt ` …