Speech as a biomarker for supported diagnosis of major depressive disorder using self-supervised representations
Yunhan Lin, Biman Najika Liyanage, Chuan Shi, Zhengwen Zhu, Yutao Sun, Yaoyao Sun, Ziwei Sunny Zhang, Jun Yang, Chenyang Xu, Yundan Liao, Yanbao Tao, Zongfeng Li, Guorui Zhao, Weihua Yue · 2026
Read Speech as a biomarker for supported diagnosis of major depressive disorder using self-supervised representations by Yunhan Lin, Biman Najika Liyanage, Chuan Shi, Zhengwen Zhu, Yutao Sun, Yaoyao Sun, Ziwei Sunny Zhang, Jun Yang, Chenyang Xu, Yundan Liao, Yanbao Tao, Zongfeng Li, Guorui Zhao, Weihua Yue (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/PMC13444143.txt.
Description
Yunhan Lin and colleagues read depression in the voice with a self-taught speech model in a 2026 study. Major depression lacks firm lab marks, so clinics lean on talks and scales. Their cohort held 1,816 people, split 910 patients and 906 healthy peers, with 23,608 standard speech clips in all.
A deep frame learned from raw speech through self-supervised steps, set against named voice models and classic feature kits. Inside checks gave an AUC mark of 0.932 for the new frame, well past classic routes. Outside checks at fresh sites held an AUC mark of 0.879 in the report. Voice traits numbered 6,373 in the classic kit, yet learned marks won. The authors pitch voice as a fast, cheap aid for depression screens.
Catalog record
| CREATOR | Yunhan Lin, Biman Najika Liyanage, Chuan Shi, Zhengwen Zhu, Yutao Sun, Yaoyao Sun, Ziwei Sunny Zhang, Jun Yang, Chenyang Xu, Yundan Liao, Yanbao Tao, Zongfeng Li, Guorui Zhao, Weihua Yue |
|---|---|
| YEAR | 2026 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 57865 |
| LICENSE | https://creativecommons.org/licenses/by/4.0 |
| RIGHTS | CC BY |
| SOURCE | Europe PMC |