The Fouriesolator: a speculative additive synth module

There’s a really neat feature of “signal” in the universe. Which is that, for any signal, any shape, any wave, can be described with the right number of sine waves played over each other.

Any signal.

If a million flutists play each sustain a single, exactly correct note over each other, the sine waves — the oooooooooo sound of their flutes, each at a different pitch and volume — then, when you listened to the million flutists, you could hear an entire Star Wars Rebel March, complete with brass, drums, and the tapping baton at the beginning.

Those flutes, stricltly speaking, would have to play nonstop for the length of the symphony, and some of them would have to be very bass. Some sort of infrasubcontrabass flutes.

The Fouriesolator is a synthesizer I’m designing with Septic Snake with the intention of eventually manufacturing the thing. So, ignoring the specifics of how it works for the time being, this is what it does.

First, it listens to a signal. Let’s say it’s a sample of Max Headroom.

For the sake of the argument, we grab one second of it by sending the Fouriesolator a one second square wave GATE signal.

Then, you know how a graphic equalizer lights up to show you a band of frequencies?

The Fouriesolator lights up each of its frequency bands according to the frequencies it heard while the gate was open. If you slide the faders each down to where it just begins to dim, then it’s set to where the frequency was in the source sound.

It is now producing a sine wave at each of those frequencies, producing a low fidelity impression of the original sound. In the illustrated case above, the time the gate was open is considered the fundamental. Each of the sliders is a proportion of that. Here, they’re a Fibonacci sequence, but maybe the curve is best as a strict exponential curve. It might be best to choose the curve you’re using for artistic purposes while you’re playing. I don’t know yet.

And finally, you can now play other sounds through it. The Fourier transformation that it’s doing is one you can apply to other sounds. So now, if I run more audio into it — say, another person’s voice — the levels of the frequency bands of the incoming signal are used to play sine waves in the appropriate frequencies of the oscillators. If I’m right, it will sound faintly like Max Headroom. And if I play a violin as the initial sound, speaking through it will sound like a speaking violin.

Hey, I make my money as an artist and a teacher. Bump my Patreon so I can continue to make art and eat food!

In addition to the summed output, each frequency sends an output of that frequency at its current amplitude. That amplitude is further modified by a control voltage output. That means that another such module can be sending its outputs to modify the output according to other parameters happening elsewhere in the modular rack.

I’m kind of into the idea of producing one that contains the minimum number of faders that are practical, then designing additional banks to add more frequency bands according to your design as you build a rack.

This is a big module. But it can do all sorts of sound. There’s a lot of experimentation to be done! I’m hoping to be able to prototype it in VCVRack or perhaps even Akso to really try it out and see how it works as an instrument.

When a tribe in the Fifth World has a particularly effective, prosocial technology, how to they ensure other tribes can do it, too?

Leave a Reply