Dual-branch hierarchical fusion network for voice-based classification of type 2 diabetes mellitus using multi-vowel acoustic features
Pengxu Jiang, Wenke Zhang, Aiqin Li, Peng Li · 2026
Read Dual-branch hierarchical fusion network for voice-based classification of type 2 diabetes mellitus using multi-vowel acoustic features by Pengxu Jiang, Wenke Zhang, Aiqin Li, Peng Li (2026) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/pmc-13582493.txt.
Description
Pengxu Jiang, Wenke Zhang, Aiqin Li, and Peng Li test whether the voice can betray type 2 diabetes. Speaking traits shift subtly with the disease, but earlier studies used thin recordings and simple features. The team recorded 378 people, patients and healthy controls, each sustaining six Mandarin vowels, and drew three families of sound features from the clips.
Their model has two branches and fuses the features in layers, with attention that leans on whichever features matter most for the disease. It sorted patients from controls with about eighty percent accuracy. The authors say the result shows that held vowels carry real signals of diabetes status. They also caution that the method needs testing on fresh groups of people before any clinic could use it.
Catalog record
| CREATOR | Pengxu Jiang, Wenke Zhang, Aiqin Li, Peng Li |
|---|---|
| YEAR | 2026 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 41 KB |
| VERSION | 1.0 |
| LICENSE | CC BY 4.0 |
| RIGHTS | CC BY |
| SOURCE | Europe PMC |