EUVIMEDEUVIMED entdecken
Open Access · CC BY 4.0

Task-Aligned Speech Representations for Intelligibility Classification in Patients with Cleft Palate

Xiao Luo, Shan Liao iD, Xue Gong iD, Zhenyu Liu, Xueni Zhao, Chenghao Li, Ling He iD, Heng Yin

Sensors · 04.11.2026 · 26 (21) · 6975

DOI: 10.3390/s26216975 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Sensors, online erschienen am 04.11.2026. Verfasst von Xiao Luo und 7 weiteren Autor:innen. Beteiligte Einrichtung: State Key Laboratory of Oral Diseases & National Clinical Research Center for Oral Diseases & Department of Cleft Lip and Palate Surgery, West China …. Hauptthemen laut MEDLINE: Cleft Palate, Speech Intelligibility, Speech. Die Arbeit nennt 30 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: the Clinical Research Program of West China Hospital of Stomatology, Sichuan University, Sichuan University collaborative research project, Clinical Research Program of West China Hospital of Stomatology, Sichuan University.

AbstractSchlagwörterBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Perceptual evaluation remains the clinical standard for postoperative cleft-palate speech, but objective patient-level assessment of overall speech intelligibility remains limited. We developed a patient-level three-class classification model using frozen Chinese HuBERT representations, principal component analysis (PCA), and regularized multinomial logistic regression. The study included 123 patients with cleft palate, including 100 in the development cohort and 23 in the held-out cohort. Each patient completed 27 standardized articulation tasks. HuBERT embeddings from each task were summarized by temporal mean and standard deviation and concatenated in a fixed clinical order to form a patient-level representation. Model development used patient-disjoint nested cross-validation followed by locked held-out evaluation. The primary model achieved 90.00% accuracy and a macro-F1 of 0.8659 in development and 86.96% accuracy, a macro-F1 of 0.8227, 93.75% balanced accuracy, and a quadratic-weighted kappa of 0.9043 in the held-out cohort. All three held-out errors involved patients with Poor intelligibility who were misclassified as Reduced intelligibility, with no direct Normal-to-Poor errors. Comparisons across six classifiers and four pretrained encoders showed no significant paired differences after Holm correction. Task-replacement analysis revealed heterogeneous reliance on individual articulation tasks. These findings support the feasibility of objective patient-level speech intelligibility assessment from standardized cleft-palate speech recordings.

Schlagwörter

MeSH-Schlagwörter (MEDLINE)

Cleft PalateSpeech IntelligibilitySpeechHumansFemaleMalePrincipal Component AnalysisClassification AlgorithmsChild

Stichwörter der Autor:innen

HuBERTacoustic sensingcleft palatepathological speechself-supervised learningspeech intelligibility

Autor:innen und Institutionen

  • State Key Laboratory of Oral Diseases & National Clinical Research Center for Oral Diseases & Department of Cleft Lip and Palate Surgery, West China School of Stomatology, Sichuan University, Chengdu 610041, China.
  • College of Biomedical Engineering, Sichuan University, Chengdu 610065, China.
  • Department of Otorhinolaryngology, Head and Neck Surgery, The Chinese University of Hong Kong, Shatin, New Territories, Hong Kong SAR 999077, China.

ORCID-Profile: Shan Liao, Xue Gong, Ling He

Förderung

  • the Clinical Research Program of West China Hospital of Stomatology, Sichuan University
  • Sichuan University collaborative research project
  • Clinical Research Program of West China Hospital of Stomatology, Sichuan University

Zitieren

APA

Luo, X., Liao, S., Gong, X., Liu, Z., Zhao, X., Li, C., He, L., & Yin, H. (2026). Task-Aligned Speech Representations for Intelligibility Classification in Patients with Cleft Palate. Sensors, 26(21), 6975. https://doi.org/10.3390/s26216975

Vancouver

Luo X, Liao S, Gong X, Liu Z, Zhao X, Li C, et al. Task-Aligned Speech Representations for Intelligibility Classification in Patients with Cleft Palate. Sensors. 2026;26(21):6975. doi:10.3390/s26216975

BibTeX
@article{Luo2026,
  author = {Luo, Xiao and Liao, Shan and Gong, Xue and Liu, Zhenyu and Zhao, Xueni and Li, Chenghao and He, Ling and Yin, Heng},
  title = {Task-Aligned Speech Representations for Intelligibility Classification in Patients with Cleft Palate},
  journal = {Sensors},
  year = {2026},
  volume = {26},
  number = {21},
  pages = {6975},
  doi = {10.3390/s26216975}
}
Kostenlos starten →