LiveSet · the synth
The synth
Most of a live electronic set is synthesis, so the synth inside LiveSet had to be an instrument rather than a token preset-player. It has ten sound engines, a voice you build from components, and a filter that morphs between shapes. It ships with 534 factory presets, and every one of them can be taken apart.
Ten engines under one voice
Pick an engine per component, or blend them. Each one is a different way of getting from nothing to a note.
Analog
A subtractive oscillator pair plus a sub, band-limited saw, square and pulse with PolyBLEP anti-aliasing, the bread-and-butter body that stays clean up high.
Wavetable
A scanned table you move by hand or with a modulator, band-limited so a fast sweep does not alias. Warp the shape as you go: bend, fold or PWM.
FM
An operator stack, operators modulating operators, for metallic bells, glassy keys and the bite techno leads live on.
Noise
Noise into a resonant bandpass into a transient hit, where snare buzz, hat sizzle and claps come from.
Physical
Karplus-Strong physical-model plucks, string, mallet, kalimba, that behave like something you struck.
Additive
A partial bank you build a tone from and then move, for organs, glassy pads and spectral risers.
Modal
A bank of tuned resonators for bells, mallets and struck bodies with a real sense of material.
Transient
A deterministic one-shot, click, impulse, noise burst or pitched blip, for the attack that has to land in exactly the same place every time.
Sample
An asset-backed sample source: a vocal chop, a resampled bass, a one-shot you brought with you.
Granular
An asset-backed granular source for evolving textures and stretched pads.
The voice
Building a voice
A LiveSet voice is a stack of components. Each component has its own engine, its own amplitude envelope, its own low-pass filter and, if you want it, its own filter envelope and drive. Layer a sub, a body and a bit of noise, and each one moves on its own terms.
Over the whole voice sits a five-mode morphing filter — low-pass, band-pass, high-pass, notch and peak, with its own drive. Morph between the modes and the character of the sound turns with it.
Character in the voice
The dirt and movement that make a sound yours are built into the voice from the start.
Unison spread
Stack detuned copies of the analog or wavetable oscillator for the wide, moving supersaw a big room asks for: a spread on the voice.
Ring modulation
Multiply two sources for clangorous, inharmonic tones.
Hard sync
Slave one oscillator to another for that tearing, aggressive sweep.
PWM
Move the pulse width for hollow, breathing timbres.
Formant
Vowel-like resonances that make a sound almost speak.
Waveshaper
Soft-clip, hard-clip, fold, bit-crush or rectify, drive with a character you choose.
Oscillator warp
Bend the raw shape before anything else touches it.
Made to move
Modulation
Every voice has an ADSR and a dedicated pitch envelope, plus glide between notes. From there, dozens of the synth's parameters are live modulation destinations, wire an LFO, an envelope, a macro or a random source to any of them and the sound breathes while it plays.
A voice, roughly
component 1 · analog saw → LP filter → env
component 2 · sub sine → env
component 3 · noise → bandpass → transient
↳ morph filter (LP↔BP↔HP↔notch↔peak) + drive
↳ ADSR + pitch env + glide
↳ dozens of live modulation destinations
Where it stands
What the engine does, and where you reach it
The synthesis itself, meaning the ten engines, the component voice, the morph filter and the modulation, is built and shipping. Some of that depth is still reached through your controller, because there is not yet a knob on screen for every parameter: envelope and LFO shaping happens on the controller instead of in a visual editor, and wavetables are one- dimensional. The sound is all here. The last of the interface is still being built.