Synheart Emotion: Privacy-Preserving On-Device Emotion Recognition from Biosignals

TL;DR

This paper evaluates machine learning models for on-device emotion recognition using wrist biosignals, demonstrating that classical ensemble methods outperform deep learning models in accuracy and efficiency, enabling privacy-preserving real-time applications.

Contribution

It systematically compares various ML architectures for on-device emotion recognition and shows classical ensemble methods are most effective for small physiological datasets.

Findings

Classical ensemble methods outperform deep neural networks and transformers in accuracy.

The optimized ExtraTrees model achieves a 4.08 MB footprint and 0.05 ms inference latency.

ONNX optimization reduces storage by 30.5% and increases inference speed by 40.1x.

Abstract

Human-computer interaction increasingly demands systems that recognize not only explicit user inputs but also implicit emotional states. While substantial progress has been made in affective computing, most emotion recognition systems rely on cloud-based inference, introducing privacy vulnerabilities and latency constraints unsuitable for real-time applications. This work presents a comprehensive evaluation of machine learning architectures for on-device emotion recognition from wrist-based photoplethysmography (PPG), systematically comparing different models spanning classical ensemble methods, deep neural networks, and transformers on the WESAD stress detection dataset. Results demonstrate that classical ensemble methods substantially outperform deep learning on small physiological datasets, with ExtraTrees achieving F1 = 0.826 on combined features and F1 = 0.623 on wrist-only features, compared to transformers achieving only F1 = 0.509-0.577. We deploy the wrist-only ExtraTrees model optimized via ONNX conversion, achieving a 4.08 MB footprint, 0.05 ms inference latency, and 152x speedup over the original implementation. Furthermore, ONNX optimization yields a 30.5% average storage reduction and 40.1x inference speedup, highlighting the feasibility of privacy-preserving on-device emotion recognition for real-world wearables.