VOICESOUNDsings vowelslight / dark

VOICE

VOICE sings vowels from the keyboard: a glottal pulse (the buzz of the vocal folds) through four formant filters, the resonances that make an "ah" different from an "oo".

sings vowels

Half

SOUND

Reach it

a custom build, FELUCCA_VOICE=1 (else on MOUTH)

ALGORITHM

VOWL

Engine

VOICE

Factory sounds

4

Playing it

VOWL on MAIN moves through the vowels A, E, I, O and U, blending between neighbours. BUZZ sets how bright and pressed the voice is, from a soft, open tone to a buzzy one. BRTH mixes in breath noise; at the top the voicing fades out and the voice whispers. TALK is a glide time: each note moves from VOWL to a second vowel, VOWL2, over that time, so a held chord can say "ah-oo".

VOICE is on the sidelines: SQRL-1's build leaves it out. A custom build with FELUCCA_VOICE=1 brings it back, picked on the PAIR page (hold HOME, tap GLO, KNOB 1). Without it, a VOICE sound in a project or a saved sound plays on MOUTH.

MAIN

Pages

VOWL

SHIFT moves the formants up or down in semitones: down for a bigger, darker throat, up for a smaller one.

Knob	Label	Range	Starts at
1	VOWL	0 .. 127	0
2	VOWL2	0 .. 127	0
3	TALK	1.0ms .. 10.0s	1.0ms
4	SHIFT	-12st .. +12st	0st

TONE

Q sets how narrow the formants are: high values ring, low ones blur. RAND pulls both vowels towards a random vowel chosen for each note, so a choir's voices do not all say the same thing.

Knob	Label	Range	Starts at
1	BUZZ	0% .. 100%	31%
2	BRTH	0% .. 100%	17%
3	Q	0% .. 100%	47%
4	RAND	0% .. 100%	0%

Where it comes from

firmware/src/eng_formant.c, Felucca's engine (GPL-3.0), by Leo Kuroshita, Hügelton Instruments. It follows D. H. Klatt's cascade formant synthesizer (JASA, 1980), with a Rosenberg glottal pulse and vowel formants from Hillenbrand et al. (JASA, 1995).