Laryngeal disease classification using voice data: Octave-band vs. mel-frequency filters
Jaemin Song, Hyunbum Kim, Yong Oh Lee · 2024
Read Laryngeal disease classification using voice data: Octave-band vs. mel-frequency filters by Jaemin Song, Hyunbum Kim, Yong Oh Lee (2024) 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-11667598.txt.
Description
Jaemin Song, Hyunbum Kim, and Yong Lee present a 2024 study on spotting throat disease from voice data. Throat cancer checks need a specialist, but voice tests could flag cases early with no invasive step. The team compared two ways to turn voice into features for a neural model. Voice samples came from 363 patients in four groups: healthy, cancer, benign growth, and fold paralysis.
An octave-band feature set beat the standard mel-frequency set by a wide margin. Accuracy was near 94 percent for the octave set, against about 71 percent for the standard one. Visual maps showed the winning set focused on noise above the voice formants. Some benign cases still looked like healthy voices. The authors back the method for early screening.
Catalog record
| CREATOR | Jaemin Song, Hyunbum Kim, Yong Oh Lee |
|---|---|
| YEAR | 2024 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 62 KB |
| VERSION | 1.0 |
| LICENSE | CC BY 4.0 |
| RIGHTS | CC BY |
| SOURCE | Europe PMC |