What is The Infinite Chorale?

An endless four-part chorale that composes itself in your browser, following the classical rules of voice leading — no parallel fifths or octaves, leading tones rising, sevenths resolving down — and explains every chord it chooses in plain English while you listen.

Is the music really generated live?

Yes. A functional-harmony grammar chooses each next chord and a voicing search moves the four voices under hard part-writing rules, in real time. Nothing is prerecorded; each chorale number replays identically from its share link.

You conduct it: demand a cadence, ask to be tricked with a deceptive resolution, park the phrase on a half cadence, pivot-modulate to the dominant or relative key, change key and tempo, or start a fresh chorale.

Can I use it for teaching?

It was built for that: project it in class, ask students to predict the next chord or name the cadence before the commentary does, then press 'Trick me' and discuss what a deceptive cadence did to their expectation. The share link reproduces the same chorale for the whole class.

Does it follow real part-writing rules?

Yes — the same rules taught in AP and college theory: SATB ranges and spacing, no parallel fifths or octaves, no voice overlap, doubling conventions per inversion, leading tones resolving upward, chordal sevenths resolving down by step, cadential six-four treatment, and more. The engine re-verifies its own output continuously.

The Infinite Chorale

This page writes four-part harmony, live and forever — real voice leading, real cadences, a real (sampled) piano — and explains every move as it makes it. You conduct.

Every chord obeys the rules your theory teacher grades: no parallel fifths, ever.

Tempo 66

—

The commentary

What You're Hearing

A chorale is music's clearest X-ray: four voices — soprano, alto, tenor, bass — moving together, one chord per beat, every note accountable to the others. It's how Bach taught harmony, and it's still how theory classes teach it today. This page composes one that never ends. A functional-harmony grammar decides where the music goes (tonic → predominant → dominant → tonic, with deceptions), and a voicing engine decides how each voice walks there without breaking the rules below.

The Rules It Never Breaks

No parallel fifths or octaves — between any pair of voices
No voice overlap or crossing
SATB ranges and spacing limits
Leading tones resolve upward to do
Chordal sevenths resolve down by step
No augmented melodic intervals
Doubling conventions per chord and inversion
Cadential six-four resolves over a held bass

The counter under the staff is honest bookkeeping: those are real voice-leading decisions, re-verified by the engine's own rule-checker. On the rare occasion the music paints itself into a corner and bends a style preference (never a grammar rule), the commentary confesses it.

How to Listen (and What to Try)

Watch one voice at a time. Pick the tenor and follow it for a phrase — smooth, mostly steps, the occasional purposeful leap. That's voice leading. Predict the cadence. When you hear the music gathering toward a stop, guess: authentic or half? Then check the commentary. Press "Trick me." Your ear expects home; the bass steps up instead. That betrayal-and-relief is a deceptive cadence, and once you've felt it, you'll never forget the name. Modulate. Watch the pivot line in the commentary — one chord wearing two name tags is the whole secret of smooth key changes.

For Teachers

Project it. Ask the class to name each Roman numeral before it sounds, call the cadence type before the commentary does, or spot the leading tone's resolution in real time. Then hand a student the controls and let them conduct. The share link replays the same chorale from the top, so you can send tonight's exact music home as listening homework. Pairs well with the Melodic Dictation Trainer (hear → write) and the key signature drills (the signatures you're seeing on the staff).

Frequently Asked Questions

Is this AI?

Not in the fashionable sense. It's a rule system — the same explicit, checkable rules from your harmony textbook, applied exhaustively and fast. Every decision is inspectable, which is exactly why it can explain itself.

Why does it sometimes sound old-fashioned?

Because these are the conventions of the common-practice chorale, on purpose. This is the sound the rules describe; hearing the rules as sound is the point.

Why quarter-ish chords with no rhythm?

Chorale texture isolates harmony — one chord per pulse, fermatas at phrase ends. Rhythm would blur the lesson.

Do the same chorale numbers always play the same music?

Yes — a chorale's number seeds every decision, so a shared link replays it exactly (your live conducting then takes it somewhere new).

Keep Going

Learn the vocabulary you're seeing: circle of fifths · key signatures · seventh chords by ear · chord progression generator · melodic dictation.