A 1D-CNN with advanced data augmentation for robust speech emotion recognition
Neha Chourasia, Chhattar Singh Lamba, Amit Kumar Gupta · 2026
Read A 1D-CNN with advanced data augmentation for robust speech emotion recognition by Neha Chourasia, Chhattar Singh Lamba, Amit Kumar Gupta (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/PMC13542176.txt.
Description
Neha Chourasia and colleagues present a speech tool in 2026. The tool recognises emotion in spoken voices. Machines that answer human feelings need such a tool. Older systems lean on hand-crafted features. They crumble with new speakers or noisy rooms. The new design uses a one-dimensional convolutional network.
Three sound descriptors feed the network as one sequence. They cover tone texture, signal crossings, and energy. Training data pooled four public acted-speech collections. Extra training copies came from noise, shifts, and pitch changes. Stretching added still more variety. Seven emotion classes were tested at the end. Accuracy reached about ninety-five percent on test data.
Catalog record
| CREATOR | Neha Chourasia, Chhattar Singh Lamba, Amit Kumar Gupta |
|---|---|
| YEAR | 2026 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 120440 |
| LICENSE | https://creativecommons.org/licenses/by/4.0 |
| RIGHTS | CC BY |
| SOURCE | Europe PMC |