AudioSignalProcessingForML

Code and slides of my YouTube series called "Audio Signal Proessing for Machine Learning"

This repository is a comprehensive collection of resources, code, and explanations for understanding and implementing audio signal processing techniques, with a focus on applications in machine learning. It serves as a learning guide, starting from the fundamentals of sound and waveforms and progressing to advanced feature extraction methods.

Course Structure

Foundational Concepts

1. Overview: Video | Slides
2. Sound and waveforms: Video | Slides
3. Intensity, loudness, and timbre: Video | Slides | Notebook
4. Understanding audio signals: Video | Slides

---

Feature Extraction Theory

5. Types of audio features for ML: Video | Slides
6. How to extract audio features: Video | Slides
7. Time-domain audio features: Video | Slides

---

Time-Domain Implementation

8. Implementing the amplitude envelope: Video | Notebook
9. RMS energy and zero-crossing rate: Video | Notebook

---

Frequency-Domain Concepts

10. Fourier Transform: The Intuition: Video | Slides
11. Complex numbers for audio signal processing: Video | Slides
12. Defining the Fourier transform using complex numbers: Video | Slides | Notebook
13. Discrete Fourier Transform: Video | Slides

---

Frequency-Domain Implementation

14. Extracting the Discrete Fourier Transform: Video | Notebook
15. Short-Time Fourier Transform explained easily: Video | Slides
16. Extracting Spectrograms from Audio with Python: Video | Notebook
17. Mel Spectrogram Explained Easily: Video | Slides
18. Extracting Mel Spectrograms with Python: Video | Notebook
19. MFCCs Explained Easily: Video | Slides
20. Extracting MFCCs with Python: Video | Notebook
21. Frequency-Domain Audio Features: Video | Slides
22. Implementing Band Energy Ratio from Scratch with Python: Video | Notebook
23. Spectral centroid and bandwidth: Video | Notebook

---

Audio examples

• audio_resources/: A collection of .wav files used for the examples in the notebooks.