Skip to main content
k2k audio logo k2k audio

Back to Crossmorph
Documentation tree

Spectral Vocoder

A vocoder takes the spectral envelope (the frequency-by-frequency amplitude shape) of one signal (the modulator) and applies it to another (the carrier).

Parameters

ParameterRangeDefault
Vocoder Depth0.0 – 1.01.0
Formant Shift-12.0 – 12.00.0
Output Gain0.0 – 2.01.0

Vocoder Depth — How strongly the modulator’s per-band amplitude shapes the carrier, 0–1. 0 = no vocoding (carrier passes unchanged). 1 = full vocoding (carrier’s amplitude is fully replaced by modulator’s per-band amplitude). Mid values blend a clean carrier with the vocoded version — useful for keeping the carrier’s tonal content audible while imposing the modulator’s rhythmic/dynamic shape.

Formant Shift — Pitch shift applied to the modulator’s spectral envelope before transferring it to the carrier, in semitones, −12 to +12. Negative shifts make the modulator content sound like a larger, lower-pitched source (“giant voice”); positive shifts produce smaller, higher-pitched character (“chipmunk voice”). 0 keeps the modulator’s natural formants intact.

Output Gain — Output level multiplier, 0–2. Linear gain (1.0 = unity, 2.0 = +6 dB). Compensates for the level shift introduced by vocoding.

Additional controls

Swap — When on, swaps the carrier and modulator inputs (Input A becomes modulator, Input B becomes carrier instead of the default). Saves rewiring the graph when you want to flip which input drives which.

Loop Mod — When on, the modulator input loops to match the carrier’s duration. Off plays modulator once and stops contributing after it ends. Useful when you want a short modulator phrase to vocode a longer carrier (loop on) vs. when you want them to play once together (loop off).

About Spectral Vocoder

A vocoder takes the spectral envelope (the frequency-by-frequency amplitude shape) of one signal (the modulator) and applies it to another (the carrier). The classic example: speech as modulator, sustained synth chord as carrier → “talking synthesizer.” This is the K2K version, working spectrally with per-band detection. Use it for: classic robot-vocal effects, rhythmic shaping (use a drum pattern as modulator to make any tonal source pulse with that rhythm), texture transformation (use noise as modulator for a “broken speaker” effect), or as a building block in morphing chains. Compare with Formant Morph (interpolates formants between two sources) and Spectral Convolution (applies spectral signature as a filter rather than dynamic shaping).


Generated 2026-05-05 from K2K_Dev@96730bdc by scripts/gen_lexique.py. Edit _intros/ or _overrides/, not this file.