Deep Learning For Audio With Python

Code for the "Deep Learning (for Audio) with Python" series on The Sound of AI YouTube channel.

This repository is a comprehensive collection of resources and code for understanding and implementing deep learning models for audio tasks. It serves as a practical guide, starting from the absolute basics (building neurons and backpropagation from scratch), moving to TensorFlow implementation, and culminating in building a complete Music Genre Classification system using various architectures (MLP, CNN, RNN-LSTM).

Note on Versioning

While this v2 release is fully functional and optimized for current environments, it may differ from the original version shown in the course. The codebase has been updated to reflect modern best practices (e.g. TensorFlow 2.16+, Librosa 0.11+) and improved dependency management. Consequently, the original course version has been deprecated; however, it remains available in the legacy branch for those wishing to follow the video content exactly.

Table of Contents

• Dataset Setup (GTZAN)
• Course Structure
  • 1. Fundamentals & Math
  • 2. Neural Networks from Scratch
  • 3. TensorFlow & Audio Preprocessing
  • 4. Music Genre Classification Project
  • 5. Advanced Architectures (CNN & RNN)
• How to Run the Scripts

Dataset Setup (GTZAN)

To run the music genre classification lessons (Part 4 & 5), you will need the GTZAN dataset. We provide an automated downloader to handle the acquisition, extraction, and folder organization for you.

• Quick Start: Run python dataset_downloader.py from the root directory.
• Prerequisites: Install requirements.txt.

Full Instructions: Please check the Instructions GTZAN file for detailed help using the downloader script or manual download steps.

Course Structure

Part 1: Fundamentals & Math

1. Course Overview: Video | Slides
2. AI, Machine Learning and Deep Learning: Video | Slides
3. Implementing an Artificial Neuron from Scratch: Video | Slides | Code
4. Vector and Matrix Operations: Video | Slides
5. Computation in Neural Networks: Video | Slides

Part 2: Neural Networks from Scratch

6. Implementing a Neural Network from Scratch: Video | Code
7. Training a Neural Network (Backprop & Gradient Descent): Video | Slides
8. Implementing Backpropagation from Scratch: Video | Code

Part 3: TensorFlow & Audio Preprocessing

9. Implementing a Neural Network with TensorFlow 2: Video | Code
10. Understanding Audio Data for Deep Learning: Video | Slides
11. Preprocessing Audio Data (MFCCs/Spectrograms): Video | Code

Part 4: Music Genre Classification Project (MLP)

12. Preparing the Dataset: Video | Code
13. Implementing a Neural Network for Classification: Video | Slides | Code
14. Solving Overfitting: Video | Slides | Code

Part 5: Advanced Architectures (CNN & RNN-LSTM)

15. Convolutional Neural Networks (CNN) Explained: Video | Slides
16. Implementing a CNN for Music Genre Classification: Video | Code
17. Recurrent Neural Networks (RNN) Explained: Video | Slides
18. Long Short Term Memory (LSTM) Explained: Video | Slides
19. Implementing an RNN-LSTM for Music Genre Classification: Video | Code

How to Run the Scripts

To ensure the models and scripts execute correctly, please follow these steps from your terminal:

2. Prepare the Environment (Recommended)

Before running inference, ensure you have the necessary dependencies installed:

pip install -r requirements.txt

2. Navigate to the Lesson Folder

Each class is self-contained. Move into the specific directory for the lesson you are studying:

cd class/folder/name  # Replace with the specific class directory

3. Execute the Script

Run the main script using Python:

python mlp.py  # Replace with the specific script name