Related Experiment Video
Updated: Jul 10, 2026
06:24
Author Spotlight: Advancements in the Fabrication of Synthetic Vocal Fold Models for Phonetic and Robotic Applications
Published on: January 5, 2024
1.3K
Feasibility of a Pediatric Voice Protocol for Artificial Intelligence Research
Siyu Miao1, Jinny Choi1, Laurie Russell2
1Department of Otolaryngology - Head and Neck Surgery, Hospital for Sick Children, University of Toronto, Toronto, ON, Canada.
Journal of Voice : Official Journal of the Voice Foundation
|December 2, 2025
Summary
Collecting pediatric voice data for artificial intelligence (AI) research is feasible. Tailoring methods to age-specific needs ensures high-quality data for AI applications in child healthcare.
Area of Science:
- Pediatric Healthcare Technology
- Computational Linguistics
- Bioacoustics
Background:
- Growing interest in artificial intelligence (AI) for pediatric voice analysis.
- Challenges in pediatric data collection include developmental differences affecting attention and adherence.
- Limited research on standardized methods for pediatric voice data in clinical settings.
Purpose of the Study:
- To assess the feasibility of collecting acoustic voice data in pediatric populations for AI research.
- To identify age-specific challenges and optimize data collection protocols.
- To establish methods for creating robust pediatric voice AI databases.
Main Methods:
- Prospective observational study involving voice acoustic tasks across four age categories.
- Data collected using iPads with headphones, incorporating demographic and validated questionnaires.
- Primary outcomes: protocol feasibility (task completion, headphone use, prompting, time); Secondary outcomes: acoustic measures (fundamental frequency, jitter, shimmer, HNR).
Main Results:
- 100 participants (50% female, 50% male) recruited from a tertiary otolaryngology clinic.
- 75% of participants completed the assessment within 10-25 minutes; high headphone compliance (90%).
- Younger children and males in younger groups required more prompting and were less likely to tolerate headphones; acoustic measures aligned with norms.
Conclusions:
- Feasible to collect acoustic data for pediatric AI voice research.
- Addressing age-specific needs and parental concerns is crucial for engagement and data quality.
- Findings inform improved recruitment, adherence, and data integrity for pediatric voice AI databases.
