Physiological Validation of Glottal Airflow Estimation Using Neck-Surface Accelerometry: A Comparison With...

Area of Science:

• Biomechanics of voice production
• Wearable sensor technology
• Speech signal processing

Background:

• Neck-surface accelerometers (ACC) show potential for ambulatory voice monitoring.
• ACC signals may capture both aerodynamic and vocal fold collision characteristics.
• Evaluating ACC's reflection of glottal collision dynamics is crucial.

Purpose of the Study:

• To investigate how effectively ACC reflects glottal collision dynamics.
• To compare ACC-derived metrics with electroglottography (EGG) across diverse phonations.
• To assess ACC's utility beyond simple aerodynamic analysis.

Main Methods:

• Collected synchronized ACC, microphone (MIC), and EGG signals from nine adults.
• Analyzed sustained phonations: modal, breathy, falsetto, and creaky.
• Used inverse filtering to estimate glottal airflow parameters and correlated with EGG contact measures.

Main Results:

• ACC and MIC signals showed similar trends in Acoustic Quotient (AQ) and Maximum Flow Declination Rate (MFDR).
• ACC-derived Normalized Amplitude Quotient (NAQ) and H1-H2 showed stronger correlations with EGG contact quotient (r = -0.830, r = -0.848) than MIC.
• ACC metrics demonstrated higher explanatory power (R^2) for glottal collision dynamics.

Conclusions:

• ACC faithfully reflects vocal fold collision mechanical energy, providing a direct representation of contact dynamics.
• ACC offers superior insights into glottal collision compared to conventional acoustic signals.
• ACC is a foundational tool for biomechanical voice assessment and monitoring mechanical load in daily voice care.