Predicting Heart Activity from Speech using Data-driven and Knowledge-based features
This work addresses the problem of non-invasive heart monitoring for medical diagnosis and monitoring, but it is incremental as it builds on existing research by applying self-supervised models to a known bottleneck.
The study tackled predicting heart activity from speech by comparing self-supervised models to traditional acoustic features, finding that data-driven representations outperform acoustic methods in predicting heart activity parameters, though individual variability affects model generalizability.
Accurately predicting heart activity and other biological signals is crucial for diagnosis and monitoring. Given that speech is an outcome of multiple physiological systems, a significant body of work studied the acoustic correlates of heart activity. Recently, self-supervised models have excelled in speech-related tasks compared to traditional acoustic methods. However, the robustness of data-driven representations in predicting heart activity remained unexplored. In this study, we demonstrate that self-supervised speech models outperform acoustic features in predicting heart activity parameters. We also emphasize the impact of individual variability on model generalizability. These findings underscore the value of data-driven representations in such tasks and the need for more speech-based physiological data to mitigate speaker-related challenges.