Jazz Patterns

A monorepo of jazz practice material: static notation/audio generated with Abjad and LilyPond, plus an interactive web app. It contains three independent subprojects and a small shared helper package.

Interactive app (live): https://jazz-scales.gkt.sh/ · Slides (work in progress): https://jazz-scales.gkt.sh/slides/

jazz-patterns/
  common/            jazz_common — shared pitch-class, note-naming, and LilyPond helpers
  projects/
    scales/          jazz_scales — multi-key scale charts (forward + retrograde) and a merged book
    blues/           jazz_blues  — annotated 12-bar blues studies
    web/             interactive client-side web app (TypeScript + Vite)
    presentation/    Marp (Markdown) slide deck about the project → /slides

The Python subprojects each have their own pyproject.toml, build.sh, and CI job, and depend on jazz_common (installed from the local path). The web app is an independent TypeScript subproject (Vite) that renders and plays a JSON exported from the Python scales model. No subproject depends on another's code.

Build

Scales (per-key PDFs/MIDI/WAV + combined Jazz-Scales-Book.pdf):

bash projects/scales/build.sh

Blues (annotated studies in Bb, F, C):

bash projects/blues/build.sh

Each Python script bootstraps a local .venv, installs jazz-common and the subproject, and writes outputs to that subproject's build/ directory.

Web (interactive app — live at jazz-scales.gkt.sh): pick a key/scale, see the notation, and play it back — transport (play/pause, stop, loop), a current-note highlight, selectable instruments (including a Salamander grand piano) with per-instrument octave, an adjustable swing feel, and diatonic interval-practice patterns (thirds, fourths, …). It's an installable PWA that works offline — cache the sounds you want for no-internet use. See projects/web/README.md for the full feature list.

cd projects/web && npm install && npm run dev

The web app auto-deploys to GitHub Pages on pushes to main that touch projects/web (.github/workflows/deploy-web.yml).

See each subproject's README for requirements, options, and details:

• projects/scales/README.md
• projects/blues/README.md
• projects/web/README.md

Requirements

• Python 3.9+ and Abjad 3.31+ (scales, blues)
• LilyPond (PDF / MIDI); FluidSynth (WAV, scales only)
• macOS: brew install lilypond fluidsynth
• Ubuntu/Debian: sudo apt-get install -y lilypond fluidsynth
• Node.js 18+ (web app only)

Citation

If you find the scales work useful, please consider citing it (see projects/scales/README.md for the full entry):

Thiruvathukal, George K. “Jazz Scale Patterns with Abjad and Lilypond”. 2025. https://doi.org/10.6084/m9.figshare.29887028

Acknowledgments

I wish to acknowledge one of the great masters of jazz pedagogy, Jamey Aebersold, whose Jazz Handbook (the "red book") at https://www.jazzbooks.com/jazz/FQBK is a great inspiration.

Jazz mentors and music colleagues

I've learned so much from all of them these past few years — this work would not exist without them. Thank you.

• Jack Cassidy — my first jazz mentor (Old Town School of Folk Music).
• Lara Driscoll — my piano teacher (Loyola University Chicago).
• Victor Garcia — mentor (Loyola University Chicago); bandleader, Victor Garcia's Listening Party.
• Dongryul Lee — my composition teacher (Loyola University Chicago).
• Christopher Madsen — Director, Jazz Studies (Loyola University Chicago).
• Michael Nearpass — my current mentor (Old Town School of Folk Music).
• David B. Wetzel — teaching collaborator in digital and electronic music (Loyola University Chicago).

Loyola University Chicago colleagues

• Konstantin Läufer — Professor of Computer Science.
• Neil Klingensmith — Associate Professor of Computer Science.
• David B. Wetzel — Teaching Assistant Professor of Computer Science.

Purdue University colleagues

• Yung-Hsiang Lu — Professor of Computer Engineering.
• James C. Davis — Assistant Professor of Computer Engineering.

Argonne National Laboratory

Beyond the music — the colleagues who back my (sometimes crazy) ideas and help turn them into great research.

• Janet Knowles — Principal Software Development Specialist, ALCF; a fellow musician whose encouragement means a great deal.
• Victor Mateevitsi — Computer Scientist, ALCF; collaborator with me and my PhD student Nicholas Synovic.
• Mike Papka — Director of the Argonne Leadership Computing Facility.
• Silvio Rizzi — Computer Scientist, ALCF; my collaborator of several years on HPC education and research.
• Shilpika — my former graduate student in Software Engineering, now a computational scientist working on visualization.