Smartphone-Based Mini-Program Voice Acquisition for Remote and Real-Time Acoustic Voice Analysis.

Area of Science:

• Speech and Hearing Sciences
• Digital Health
• Acoustic Analysis

Background:

• Remote acoustic analysis of voice disorders is challenging.
• Existing methods often require specialized equipment.
• Smartphone technology offers a potential solution for accessible voice assessment.

Purpose of the Study:

• To validate a novel WeChat mini-program, "Voice Acquisition," for smartphone-based acoustic voice analysis.
• To compare the reliability of Voice Acquisition against the conventional Computerized Speech Lab (CSL) software.
• To assess the utility of Voice Acquisition for screening and follow-up of voice disorders.

Main Methods:

• Developed and utilized the "Voice Acquisition" WeChat mini-program on smartphones.
• Collected sustained vowel /a/ phonations from 77 patients and 80 healthy volunteers.
• Simultaneously recorded data using Voice Acquisition and CSL software in both soundproof and regular environments.

Main Results:

• Voice Acquisition demonstrated moderate to strong correlations (r = 0.543-0.989) with CSL across most acoustic parameters.
• Maximum phonatory time (MPT) and fundamental frequency (F0) mean showed high stability and correlation.
• The mini-program's acoustic parameters, such as the cepstral/spectral index of dysphonia (CSID), effectively discriminated between pathological and normal voice groups.

Conclusions:

• The "Voice Acquisition" mini-program is a reliable tool for remote voice data collection, showing strong correlation with CSL.
• Preliminary findings suggest its potential for screening voice disorders.
• Further validation is required to establish its definitive clinical utility for voice disorder management.