Audio Synthesizer

Composition Gallery
, , , , , , , ,
1

Running through some basic audio synthesis concepts with Vuo audio tools.

Since there are such amazing modular environments out there solely dedicated to audio, I hadn’t given Vuo much thought for audio. Vuo of course has its own special charm, and having made this comp I can now better imagine some interesting results integrating some audio with visuals in Vuo.

I got this to the point of getting the concepts up and running smoothly with (I hope) no stray events, but I haven’t done much about streamlining. As it is, it is just under the threshold for maxing out my 2020 Mac Mini M1 running macOS Big Sur 11.2.3 and Vuo 2.3.0 (maxed out becomes sluggish with pulsing drop outs). All the duplicate downstream processes here would be no problem for dedicated audio software. Vuo can hang in there but feels fragile for real-time tweaking.

The settings in the comp are where I left off in “testing all parameters” mode. When running the comp you should hear two main sound sections, a fast moving, sweeping sequence (the original idea and the bulk of the comp) and a slower moving low supporting section. Some parameters are published, many/most are part of subcomps. The subcomps add up to some basic audio tools, with parameters/faders calibrated for audio.

Concepts explored (roughly in order of signal flow):

  • Clock: BPM + clock divisions.

  • Sequencer (with many modes and scales).

  • Portamento (glide).

  • Vibrato (pitch modulation).

  • Sound generators usingMake Audio Wave for “oscillators” and Make Random List + Split Audio by Frequency for filtered noise.

  • Vuo’s ring modulator.

  • Stereo mixing and pan.

  • Waveshaping (a “classic” sine wave wavefolder and tanh saturation).

  • Audio filters using Split Audio by Frequency(LP, HP, BP and Notch, set up in a “state variable filter” style).

  • “VCA” section – volume, tremolo (AM/amplitude modulation), enveloping (using the Curve and Schedule nodes as a basis for making a multi-stage envelope, in this case triggered Attack-Hold-Decay envelopes).

  • Modulation using different methods.

  • Three analog-ish “analyzer” visualizers for the stereo signal (nothing too fancy cuz visuals compete with audio for resources): oscilloscope-style waveforms for L/R channels (using Make Waveform Image), sequencer visualizer, level meters using Calculate Loudness. The noise parameter of Add Analog Image Distortionresponds to the noise oscillator.

This replaces the previous AudioSynth composition. (By the way, if it overloads your system, start by disabling the low drone section, quit other apps, browsers, etc.)

Edit: experiencing a bit of “ignorance is bliss” – having made this just to see what I could get away with in Vuo, I am now backtracking through old discussions about audio. Folks have been discussing Vuo audio for a while, for example: https://community.vuo.org/t/-/5258

AudioSynth.v0.3.zip (48.3 KB)

AudioSynth.png

5 Likes
2

Looks like amazing work but I can’t hear anything. Am I supposed to send it midi? I pressed the mouse and keyboard.

3

Should just work when you push Run, no MIDI required. Does not require any mouse/keyboard input, sounds only change by twiddling parameters in subcomps and published ports. What about your audio output device selected in Send Live Audio?

(In fact, it would be a good idea to add some basic input from mouse/keyboard… And/or MIDI CC’s maybe, but a playable MIDI instrument is another project).  

4

Hi yes you were correct the “Send live audio” is set to headphones which I didn’t have plugged in. So I can hear audio but actually it still doesn’t work as intended because it’s glitchy in a way consistent with cpu/gpu overload. Or there could be some other problem in there but it’s such a huge compilation I don’t have time to really look for issues. I like the many audio nodes you made however I think I can imagine using some of them.  

5

Yes, very heavy on resources. You could try disabling the part above the main section, the low droney part, and/or the visuals.