From the physics of digital sound to AI-assisted feature extraction and clustering. The companion guide Introduction to Digital Audio in Python explains the foundations.
Prerequisites: Module 00B (Python Fundamentals)
Estimated time: ~8–12 hours
After this module you can turn audio into features and embeddings, batch-process folders of files, and cluster sounds.
| # | Notebook | What it covers |
|---|---|---|
| 00 | Foundation - Understanding Digital Audio | Building audio from numbers, microphone capture, file I/O, waveforms, and spectrograms |
| 01 | Feature Extraction from Digital Audio | DFT, spectrograms, tempo and beat, musical meter, chords, melody (CREPE-based pitch tracking) |
| 01B | Music Feature Extraction (AI/ML) | Extracting learned audio embeddings from music tracks |
| 02 | Feature Extraction - Batch Process | Running feature extraction across folders of audio files |
| 03 | Digital Audio Clustering | Embeddings to k-means/HDBSCAN with PCA and UMAP visualization |
Content in this module¶
- Sound as Digital Audio Data - An Introduction
- Understanding & Exploring Digital Audio as Data
- Exploring Features of Musical Tracks with Python
- Using AI/ML for Feature Extraction from Musical Tracks
- Feature Extraction from Digital Audio with Batch Processing
- From Audio Tracks to Acoustic Embeddings and Clusters