The Language of Sound: a six-part series on how music is actually built

Watch the series: The Language of Sound. Licensed CC BY 4.0.

Two instruments play the same note. Why don't they sound the same?

I kept wanting a version of music education that did not put a person at a whiteboard. Just the sound itself, drawn as you hear it. So I built one. The Language of Sound is a six-part series on how music is actually built, from the smallest unit, a single waveform, all the way up to how a machine writes a song on its own. The written map behind it, every altitude named, is the field guide The Language of Sound.

The idea

I wanted it to feel like a whole thing, start to finish. So I thought of it the way you would a book, or a season of television. One complete arc you could hand someone and say watch this in order and you will get it. Season One is that arc, and it is finished. Six episodes, each building on the last. It can always go deeper, and it will. But the first season is a closed loop, a complete work, not an open-ended drip.

Season One

1. The Note, waveform, timbre, harmonics, and the envelope that gives a sound its life.
2. The Instrument, layering and synthesis, giving a wave a material, an action, and a size.
3. The Arrangement, key, chords, progression, tension and release.
4. The Mix, pan, depth, tone, and width.
5. The Master, loudness, glue, and measuring instead of guessing.
6. How the Machine Thinks, the rhythm and melody behind the numbers.

New episodes twice a week. The first two are live now.

How it is made

There is no teacher on camera and no characters. Everything you see is driven by the real audio. A live oscilloscope traces each waveform as the Governor names it, a harmonic ladder climbs as a filter opens, and the drops draw the actual summed track in four colors: four voices, one tune, made visible. Der Gouverneur, a deep Alpine philosopher voice, narrates the body of each episode. I host the open and the close.

The whole thing runs on a laptop. No GPU, no subscriptions. Animation in Python and PIL, audio in ChipForge, the numpy synthesizer that also writes the teaching tracks, narration through ElevenLabs, assembled with FFmpeg. The music for each episode is built from the very technique that episode explains, so the lesson and the song are the same thing.

License

Films: CC BY 4.0. Remix it, repost it, drop it into your own thing. Credit "Napkin Films / Organic Arts LLC" and link CC BY 4.0. Engine code (Napkin Films, ChipForge) is GPL-3.0-or-later. ElevenLabs voice audio is licensed content and is not redistributed.