Synheart Emotion: Privacy-Preserving On-Device Emotion Recognition from Biosignals
Henok Ademtew, Israel Goytom · 2025
Read Synheart Emotion: Privacy-Preserving On-Device Emotion Recognition from Biosignals by Henok Ademtew, Israel Goytom (2025) 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/arxiv-2511-06231.txt.
Description
Henok Ademtew and Israel Goytom test whether emotion recognition can run privately on a wrist device instead of in the cloud. Cloud inference exposes sensitive body data and adds delay, both bad fits for real-time wearables. Their study compares model families on heart-related signals from a wrist sensor, using the WESAD stress dataset.
Classical ensemble methods clearly beat deep networks and transformers on this small body-signal dataset. The best method, ExtraTrees, scored far higher on combined features and still led on wrist-only features. Converted to an efficient format, the wrist model shrank to about four megabytes with very fast response, showing that private on-device emotion sensing is practical today.
Catalog record
| CREATOR | Henok Ademtew, Israel Goytom |
|---|---|
| YEAR | 2025 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 42 KB |
| VERSION | 1.0 |
| LICENSE | http://creativecommons.org/licenses/by/4.0/ |
| RIGHTS | CC BY |
| SOURCE | arXiv |
Source: https://arxiv.org/abs/2511-06231