Music-Classification

Overview

Music Genre Classification using a Convolutional Neural Network (CNN).

---

Project Structure

• Dataset: Place the dataset in the following structure:

  dataset/
  ├── data/
  │   └── genres_original/
  

• Metadata: The file project_data/tracks.csv contains the list of all tracks in the dataset and their names.
• Spectrograms: Spectrograms will be generated and stored in the project_data/spectrograms/ folder.

---

Setup Instructions

1. Install Poetry

Poetry is used for dependency management. Install it using the following command:

pip install poetry

2. Install Dependencies

Run the following command to install all dependencies:

poetry install

3. Activate the Virtual Environment

Select the created virtual environment in your IDE or activate it using:

poetry shell

4. Add Dependencies

To add a new dependency to the project, use the following command:

poetry add <package-name>

For example, to add numpy:

poetry add numpy

---

Usage

1. Extract Spectrograms

Spectrogram extraction is done using the audio_to_spectrogram_v2.ipynb notebook. The notebook contains the code to extract spectrograms from the audio files in the dataset.
The spectrograms will be saved in the project_data/spectrograms/ folder.

2. Load Individual Spectrograms

Use the utility function from utils.py to load individual spectrograms. Pass the path to the spectrograms folder and the track name as arguments.

---

Notes

• Ensure the dataset is correctly placed in the dataset/ folder before running any scripts.
• For more details on the dataset, refer to the fma_metadata documentation.

---