Categories
Synthesis Writing

“Where’s the Keyboard?” Synth/Voice – synthesizer as audible voltage sensor

At the beginning of this year I demonstrated the basics of sound synthesis to a group of students with my Moog Werkstatt 01 semi modular synth and an oscilloscope. The demo was part of a Drawing Week workshop that I called Sound, Signal, Trace and taught with Peter Blamey at NAS (thanks Peter!). There were lots of great questions during and after the demo including the necessary ones that I should have foreseen like “where’s the keyboard?”

“A synthesizer is not a keyboard.” I should lead with that next time. It made me think, though, about how far from a keyboard a synth can be. At the non-keyboard end of the spectrum we might treat the synthesizer not as something to be played but as an instrument that can sing with its own voice. No keyboard, no sequencer, the bare minimum patch cables.

The synthesizer is capable of sensing and making audible tiny fluctuations in voltage, a fact that made early oscillators notoriously difficult to keep in tune. This was a problem for the synth as keyboard and even temperament tuning. Not a problem if you see unstable oscillator pitch as a means for the synth to sing.

While the tuning on most analogue oscillators is now very stable they still need to warm up before they hold a consistent pitch. That warm up period is a beautiful time if, like me, you are interested in exploring the idiosyncrasies of analogue circuitry, in hearing synth circuits assert their agency by singing their songs.

What is this synth song that you speak of?

A synth song happens when we feed it parameters and it gives us back so much more in return. We can tell a synth what to do with keyboards, sequencers, envelopes, LFOs, VCAs, random voltages, accelerometer data, etc. When we do this we feed it parameters and it gives us back more or less what we expect. We can also invite a synth circuits to perform their own fluctuations. Frequency modulation, from extremely slow oscillator drift to FM synthesis (audio rate), is an exquisite technique for such performances by synth circuitry.

Different oscillator circuits, in different cases, with different power supplies, in warm or cold weather, will take different amounts of time to warm up. Use one of those oscillators to modulate the pitch of another (FM) and what you hear are precisely those differences. Self-oscillating filters may never fully stabilise in pitch. If we sample and hold a voltage, and use it to control the pitch of an oscillator it will rarely if ever be perfectly held. This ‘voltage droop’ might be considered a technical deficiency. We can also think of it as a nice, slow modulation source.

A synth can sense and make audible exactly this kind of slow voltage change. This is the synthesizer as an audible voltage sensor, something I talk more about in the companion essay to this piece Slowfast Synthesis: The Earth’s Atmosphere as Modulation Source.

This project has been supported by the Australian Government through the Australia Council, its arts funding and advisory body.