You are viewing the SNL 2018 Archive Website. For the latest information, see the Current Website.

Poster D10, Friday, August 17, 4:45 – 6:30 pm, Room 2000AB

Delayed auditory feedback, vocal oscillations, and embedded rhythms

François-Xavier Brajot1;1Ohio University

Introduction: Corrections in on-going speech production for changes in sensory feedback suggest that sensorimotor loops act as negative feedback systems. An important property of negative feedback systems is that they tend to oscillate when delay is introduced into the system. Delayed auditory feedback should therefore promote oscillations in speech. Methods: 20 young adults sustained a vowel and repeated syllables as auditory feedback was delayed from 0 to 600 ms. Vocal fundamental frequency, intensity and formant frequencies were extracted from the audio. Fourier Transforms were carried out on each of the measures and the largest spectral peak was retained to model the primary oscillation in the signal. Results: Delayed auditory feedback enhanced a low-frequency oscillation of fundamental frequency and intensity. The period of the oscillation increased to over 2 seconds and peak-to-peak amplitude by over 20 Hz as delay magnitude increased to 600 ms. Fluctuations were also observed in formant frequencies, but were not consistently periodic in nature. Linear mixed-effects modeling confirmed that delay was a significant predictor of the frequency of oscillation for fundamental frequency and intensity, but not formant frequencies. The effect was observed in syllable repetitions as well and correlated with a decrease in syllable-based spectral energy. Conclusion: The presence of spectral energy at low frequencies under normal conditions suggests that this delay-induced oscillation is an enhancement of an existing fluctuation in the voice. The differential effect between vocal (fundamental frequency, intensity) and articulatory measures (formant frequencies) suggests that the laryngeal system is particularly susceptible to the effect, which in turn alters the coupling between phonatory and articulatory systems during speech. Future directions include applying non-linear signal decomposition methods to determine whether this feedback-dependent fluctuation represents a basic physiological rhythm embedded in the speech signal important for articulatory-prosodic coordination of speech.

Topic Area: Speech Motor Control and Sensorimotor Integration