This repository contains the resources for the Generative Music AI Course on The Sound of AI YouTube channel.
Content is organised per lecture. Each lecture has a dedicated folder with the following structure:
- Slides: PDF slides for the lecture (if available)
- Code: Code examples for the lecture (if available)
The course is produced by The Sound of AI in collaboration with the Music Technology Group at Universitat Pompeu Fabra, Barcelona.
There are 3 main instructors for the course:
- Valerio Velardo, founder of The Sound of AI
- Iran Roman, post-doctoral researcher at the Music and Research Lab at NYU.
- Xavier Serra, director of the Music Technology Group at Universitat Pompeu Fabra, Barcelona.
All the videos of the course are available in this YouTube playlist on The Sound of AI channel.
Join The Sound of AI Slack community to discuss the topics of the course. Use the #generative-music-ai-course channel.
After the video lectures end in the first week of December 2023, there'll be an in-person workshop. You can get more info about the event and register at this website.
The code examples in Part 2 of the course are written in Python 3.8.
Create a Python virtual environment.
Then, run the following code in the root directory:
$ pip install -r requirements.txt
- Course overview
- What's generative music?
- History of generative music
- Use cases
- Ethical implications
- Symbolic vs audio generation
- Generative techniques
- Limitations and future vision
- Generative grammars
- Chord generation with generative grammars
- Markov chains
- Melody generation with markov chains
- Cellular automata
- Drum generation with cellular automata
- Genetic algorithms
- Melody harmonization with genetic algorithms
- Transformers - Part 1
- Transformers - Part 2
- Melody generation with transformers