Jove
Visualize
Contact Us
JoVE
x logofacebook logolinkedin logoyoutube logo
ABOUT JoVE
OverviewLeadershipBlogJoVE Help Center
AUTHORS
Publishing ProcessEditorial BoardScope & PoliciesPeer ReviewFAQSubmit
LIBRARIANS
TestimonialsSubscriptionsAccessResourcesLibrary Advisory BoardFAQ
RESEARCH
JoVE JournalMethods CollectionsJoVE Encyclopedia of ExperimentsArchive
EDUCATION
JoVE CoreJoVE BusinessJoVE Science EducationJoVE Lab ManualFaculty Resource CenterFaculty Site
Terms & Conditions of Use
Privacy Policy
Policies

Related Experiment Video

Updated: Jun 23, 2026

Machine Learning-Based Cough Tone Classification: Diagnostic Exploration of Chronic Obstructive Pulmonary Disease and Respiratory Tract Infections
06:22

Machine Learning-Based Cough Tone Classification: Diagnostic Exploration of Chronic Obstructive Pulmonary Disease and Respiratory Tract Infections

Published on: September 19, 2025

Differentiating Etiologies of Dysphonia: A Machine Learning Approach With Multidomain Acoustic Features.

Yat Chun Au1, Manwa L Ng1

  • 1Speech Science Laboratory, Faculty of Education, University of Hong Kong, Hong Kong, China.

Journal of Voice : Official Journal of the Voice Foundation
|June 20, 2026
PubMed
Summary

Machine learning models can differentiate pathological voice disorders using acoustic features from connected speech. XGBoost demonstrated strong performance, aiding in the objective assessment of voice conditions.

Related Concept Videos

You might also read

Related Articles

Articles linked to this work by shared authors, journal, and citation graph.

Sort by
Same author

The impact and application exercises on vocal fatigue.

Scientific reports·2026
Same author

Automatic assessment of voice quality using machine learning.

Logopedics, phoniatrics, vocology·2026
Same author

Cervical auscultation using Doppler ultrasound for swallows of different consistencies by young and older healthy adults.

Logopedics, phoniatrics, vocology·2026
Same author

Acoustic changes in consonant production with a face maska).

The Journal of the Acoustical Society of America·2026
Same author

Vowel Production Associated with Cantonese Speakers of Parkinson's Disease.

Folia phoniatrica et logopaedica : official organ of the International Association of Logopedics and Phoniatrics (IALP)·2026
Same author

Adaptation and Validation of the Acoustic Voice Quality Index for Voice Disorder Assessment for the Mandarin Chinese-Speaking Population.

Journal of voice : official journal of the Voice Foundation·2025

Area of Science:

  • Speech science and biomedical engineering.
  • Computational linguistics and machine learning applications in healthcare.

Background:

  • Pathological voice disorders require accurate differentiation for effective treatment.
  • Current diagnostic methods may be subjective or require specialized equipment.

Purpose of the Study:

  • To develop and assess the feasibility of using machine learning (ML) for automatic classification of voice disorders (organic, functional, neurologic).
  • To identify key acoustic features in connected speech that contribute to differentiating these disorders.

Main Methods:

  • Analysis of 584 pathological voice recordings using 29 acoustic features from multiple domains.
  • Training and evaluation of Random Forest, Extreme Gradient Boosting (XGBoost), CatBoost, and an ensemble model using cross-validation.
Keywords:
CSIDCategorical classificationConnected speechMachine learningVoice disordersXGBoost

More Related Videos

Asthma Detection Research Based on Voice Signal Processing and Machine Learning
04:04

Asthma Detection Research Based on Voice Signal Processing and Machine Learning

Published on: July 22, 2025

Related Experiment Videos

Last Updated: Jun 23, 2026

Machine Learning-Based Cough Tone Classification: Diagnostic Exploration of Chronic Obstructive Pulmonary Disease and Respiratory Tract Infections
06:22

Machine Learning-Based Cough Tone Classification: Diagnostic Exploration of Chronic Obstructive Pulmonary Disease and Respiratory Tract Infections

Published on: September 19, 2025

Asthma Detection Research Based on Voice Signal Processing and Machine Learning
04:04

Asthma Detection Research Based on Voice Signal Processing and Machine Learning

Published on: July 22, 2025

  • Assessment of performance using accuracy, precision, recall, specificity, and F1 scores, with feature importance analysis for interpretability.

    • Main Results:

    • XGBoost achieved the highest performance (accuracy=0.82, F1=0.73), outperforming other models.
    • Organic dysphonia had the highest F1 score (0.86-0.87), while functional dysphonia was most challenging (F1=0.42-0.60).
    • Cepstral-spectral measures, such as cepstral peak prominence smoothed and harmonics-to-noise ratio, were dominant predictors.

    Conclusions:

    • Gradient-boosting ML, particularly XGBoost, effectively differentiates voice disorder etiologies using connected speech acoustics.
    • Cepstral-spectral features offer physiologically plausible markers for objective voice disorder assessment.
    • ML-assisted acoustics can supplement existing evaluations, though improving functional dysphonia detection requires further research.