Delayed Auditory Feedback Fluency Simulator
Interactive 3D model of delayed auditory feedback (DAF): a phase-locked speech oscillator receives its own voice back after a set delay, showing exactly why fluency breaks down near 150-200ms and how masking noise and feedback gain change the disruption clinicians see in DAF therapy for stuttering.
Speech-language pathologists use Delayed Auditory Feedback (DAF) — replaying a speaker's own voice back to them a fraction of a second late — as an established fluency-shaping technique for stuttering. This simulator models speech timing as a phase-locked oscillator that reads its own output back through a delay line, exactly like a DAF headset, and shows why the resulting disruption to fluency is worst around a delay of roughly 150-200 ms rather than growing steadily with delay. Adjust the feedback delay, coupling gain and masking-noise level to see the syllable ring desynchronise, the produced rate drift, and a live disfluency index (the coefficient of variation of the inter-syllable interval) rise and fall exactly as reported in the classic Fairbanks delay-response curve.
A phase-locked speech oscillator receives its own voice back through a delay line, showing why Delayed Auditory Feedback (DAF) disrupts fluency most around 150-200ms and how gain and masking noise change the disruption clinicians use for stuttering therapy.
3D · Three.js / WebGL renderer · 60 FPS target · runs fully client-side, no install