Deep Learning Classification of Multicategory Voice Disorders Using Multi-Vowel Mel Spectrograms
A deep learning model using multi-vowel Mel spectrograms can non-invasively classify common voice disorders with clinically meaningful accuracy and shows particular promise as a screening tool for high-risk laryngeal lesions.
Authors
- Yimiao Wang (ORCID: https://orcid.org/0000-0003-4619-6339)
- D. Li
- K. Sandy Pang (ORCID: https://orcid.org/0000-0002-0683-2129)
- Liang Wang (ORCID: https://orcid.org/0009-0008-6088-8589)
- Siyi Chen (ORCID: https://orcid.org/0000-0003-1844-5860)
- Peijie He
- Shuaichi Ma
- Chen Zhao
- Zhiyan Lu
Institutions
- University of Shanghai for Science and Technology (CN)
- Eye & ENT Hospital of Fudan University (CN)
Publication Details
- Journal
- Ear Nose & Throat Journal
- Published
- 2026-03-13
- DOI
- https://doi.org/10.1177/01455613261428272
- Primary Topic
- Voice and Speech Disorders
- Type
- article
- Field-Weighted Citation Impact
- 0.00