Dynamics of plosive consonants via imaging, computations, and soft electronics

Area of Science:

• Fluid dynamics
• Aerosol science
• Biomedical engineering
• Soft electronics

Background:

• Quantitative understanding of aerosol and droplet transmission during speech is limited.
• Plosive sounds in speech generate unique flow-particle dynamics crucial for transmission.
• Monitoring speech patterns can aid in understanding disease spread.

Purpose of the Study:

• To integrate flow-particle dynamics of speech production with skin-integrated electronics for sound monitoring.
• To characterize particle dynamics in speech-generated flows.
• To develop a wearable sensor system for detecting speech patterns.

Main Methods:

• Analysis of flow-particle dynamics during plosive sound production.
• Characterization of Lagrangian acceleration and pair dispersion.
• Development of a wireless, soft electronic device for motion capture.
• Application of a convolutional neural network for plosive sound detection.

Main Results:

• Identification of diffusive and ballistic regimes in particle transmission, separated by particle size.
• Characterization of Lagrangian particle dynamics, showing isotropic turbulence in the diffusive regime.
• Successful detection of plosive sounds in multiple languages using the developed sensor and AI.

Conclusions:

• Skin-interfaced wireless sensors can continuously measure critical speech patterns based on neck biomechanics.
• The developed soft electronic device enables detection of plosive sounds, relevant for infectious disease transmission studies.
• This interdisciplinary work bridges fundamental physics with wearable technology for health monitoring.