Extreme Vocal Techniques: Mechanism & Safety

False-cord (ventricular-fold) distortion is not one sound source but two, stacked. The true vocal folds keep vibrating normally underneath. Above them, the ventricular folds are drawn inward by supraglottic constriction and set vibrating by the airstream, oscillating at a sub-multiple of the glottal frequency — ratios of 2:1, 3:1 and 4:1 have been documented, and in one subject an aperiodic, chaotic pattern with no stable ratio at all. Counterintuitively, the effect is produced by decreasing true vocal fold adduction while increasing subglottal pressure and airflow — the folds come apart, not together, which matters for teaching because squeezing harder is the wrong instruction and points toward strain rather than technique. Peer-reviewed laryngostroboscopic work separates what performers lump together as “growl” into four distinct effects by vibrating structure — distortion, growl, rattle and grunt — each mapped to a different part of the supraglottic anatomy.

Fry-initiated scream starts from the pulse register (vocal fry), where the folds are thick and slack and the defining vibratory behaviour is period doubling — pairs of cycles of unequal duration. In screaming use, that irregular low-frequency pulse train is driven at far higher airflow and pressure and combined with supraglottic narrowing, so perceived pitch and brightness rise sharply while the aperiodic noise floor is retained. “Fry scream” is a performer term, not a research category, and a common online claim that it is produced specifically by the arytenoid cartilages could not be verified in peer-reviewed sources — the verified arytenoid attributions belong to growl and rattle, not to a fry-initiated scream.

Layered / multi-tracked shout gets its “room full of people” quality from the chorus effect: multiple near-identical performed takes, each with small random deviations in onset timing and pitch, produce time-varying comb filtering and beating between partials. A duplicated (rather than re-performed) layer produces no chorusing at all unless pitch or timing is artificially varied. The individual layers are typically a projected, speech-adjacent shout with less supraglottic distortion than a growl, precisely so consonants stay legible.

Clean-to-distorted register transitions combine two things that must be understood separately. First, the underlying laryngeal mechanism can shift between modal (chest, thyroarytenoid-dominant) and falsetto (head, cricothyroid-dominant) voice, with the transition zone spanning several adjacent pitches rather than one sharp threshold. Second, and separately, trained singers can switch the supraglottic distortion source on and off largely independently of which underlying phonation type is running beneath it — that independence is the technical basis of a controlled clean-to-distorted move. Combining both findings with the growl mechanism above gives the safe design principle: reach the pressure threshold for a distortion layer by opening the glottis and raising airflow, not by increasing fold adduction. Adding distortion by squeezing, especially inside an already-unstable transition zone, is the documented failure mode.

Throat-singing-adjacent techniques actually split into two mechanically unrelated effects. A low, buzzing, subharmonic style has been shown by high-speed imaging to work by the same family of mechanism as false-cord distortion — ventricular folds vibrating at exactly half the true-fold frequency, closing over every second glottal closure. A separate high, whistle-like overtone effect is not a second sound source at all: vocal-tract shaping merges two formants into one narrow, high-Q filter peak, hugely amplifying a single harmonic of an ordinary glottal tone — a linear filtering phenomenon, not a nonlinear source phenomenon, and mechanically closer to ordinary phonation than to any of the distortion techniques above.

Everything above is about the mechanism and risk of a deliberate technique. Underneath that, independent of whether anyone attempts one of these techniques at all, sits ordinary vocal hygiene — worth stating because this page doesn't cover it elsewhere. The National Center for Voice and Speech recommends extra hydration around anything that dries the voice, specifically “an extra glass of water for every cup of coffee or shot of alcohol,” and, where the urge to cough or clear the throat shows up, sipping water, swallowing or a throat lozenge instead — repeated hard throat-clearing is itself a mechanical stress on the folds, on top of whatever technique caused the urge.