Source-linked AI summary
EarStreAM: A Closed-Loop Earable System for Personalized Stress-Adaptive Meditation
Jonas Hummel, Luisa Faust, Elias Müller, Eva Bertog, Valeria Zitz, Marius Johannes Prill, Luca L. Bennardo, Luisa Weber, Tobias Röddiger, Michael Beigl
TL;DR
Workplace meditation approaches may not adapt to users’ momentary physiological state, limiting support when it is needed. EarStreAM combines in-ear sensing with personalized LLM-generated meditation that adapts to physiological response. Its demonstration offers a biosignal-adaptive mode and a meditation-only mode.
Problem
Many workplace meditation approaches are not adaptive to momentary physiological state, limiting support when it is needed most.
Source: §1 · p. 1Method
EarStreAM uses in-ear physiological sensing to trigger personalized LLM-generated meditation that adapts to the user's response and ends when stress returns to baseline.
Source: §1 · p. 1, §3 · p. 2, §3.3 · p. 3Results
Takeaways & Limitations
Abstract
from arXiv · showhide
We present EarStreAM, a closed-loop earable system for stress-adaptive meditation that integrates in-ear physiological sensing with personalized, real-time intervention. Leveraging OpenEarable 2.0's multimodal sensing, EarStreAM continuously monitors physiological signals and detects elevated stress from heart rate and heart rate variability. Upon detection, the system initiates a personalized guided meditation generated by an LLM and adapted in real time to the user's stress state. The demo offers a hands-on experience of stress-adaptive meditation in two modes: a biosignal-adaptive meditation with optional stress induction to illustrate closed-loop adaptation, and a meditation-only mode focusing on EarStreAM's generative personalization capabilities. The demo highlights how in-ear sensing, closed-loop adaptation, and personalized generative meditation can be integrated into an earable system for real-time stress support in demanding office work contexts.
1 Introduction
Workplace stress is associated with reduced productivity, job satisfaction, and retention, while brief workplace meditation approaches may not adapt to momentary physiological state. EarStreAM combines in-ear sensing with personalized, stress-adaptive meditation and demonstrates it in two modes.
1 Introduction
- Workplace stress is associated with reduced productivity, lower job satisfaction, and increased turnover intentions.Prior work has explored brief meditation exercises that can be integrated into workplace routines.Source: §1 · p. 1
- Many workplace meditation approaches are not adaptive to momentary physiological state, limiting support when it is needed most.Source: §1 · p. 1
- EarStreAM combines in-ear PPG sensing with personalized LLM-generated meditation that ends when relaxation is detected.It builds on StreAM toward a self-contained system for everyday workplace use.Source: §1 · p. 1
- The demo offers biosignal-adaptive meditation, optionally after stress induction, and personalized meditation without prior induction.The modes let attendees experience closed-loop adaptation and generative personalization.Source: §1 · p. 1
2 Related Work
Prior work has explored biosignal-based stress detection and LLM-generated meditation, but closed-loop earable interventions remain comparatively underexplored. EarStreAM builds on StreAM with OpenEarable 2.0 to realize a fully closed-loop intervention using an earable and smartphone.
2.1 Biosignal-Adaptive Stress Interventions
- Closed-loop, biosignal-adaptive stress interventions remain comparatively underexplored, and no earable system had yet been demonstrated with continuous detection and automatic, real-time-adaptive intervention.Prior examples include heart-rate-guided resonant breathing and physiological adaptation of virtual environments.Source: §2.1 · pp. 1–2
2.2 LLM-Based Adaptive Meditation Systems
- Building on StreAM, EarStreAM uses OpenEarable 2.0 sensing and audio capabilities for a fully closed-loop intervention with an accompanying smartphone.The system also introduces refinements targeting robustness, personalization, and usability.Source: §2.2 · p. 2
3 EarStreAM: A Closed-Loop Earable System for Personalized Stress-Adaptive Meditation
EarStreAM pairs OpenEarable 2.0 with a smartphone to monitor in-ear physiological signals and adapt personalized meditation to the user's response. Its processing and detection pipeline establishes individualized baselines, screens for movement, and uses physiological feedback to guide intervention and recovery.
3 EarStreAM: A Closed-Loop Earable System for Personalized Stress-Adaptive Meditation
- EarStreAM monitors in-ear PPG-derived heart rate and heart rate variability, triggers personalized meditation when stress is elevated, and ends it when stress returns to baseline.The system also offers a personalized meditation-only mode.Source: §3 · p. 2, Figure 2 · p. 3
3.1 Signal Processing Pipeline
- In-ear PPG is sampled at 200 Hz; artifact removal, signal-quality gating, and RR-interval processing produce HR updates every 3 s and RMSSD when available.RR intervals accumulate in a rolling buffer of up to 5 minutes, with a 1-minute minimum.Source: §3.1 · p. 2
3.2 Context-Aware Stress Detection
- Stress detection combines demographic priors with five minutes of measurements, then flags sustained stress when HR rises and HRV falls beyond a default 10% threshold.The system reassesses after each meditation segment and uses HR alone for recovery because HRV responds more slowly; it suppresses meditation when recent movement exceeds specified activity thresholds.Source: §3.2 · pp. 2–3
3.3 Personalized Meditation Generation
- Meditation segments of approximately 280 words combine user preferences and current physiology in an LLM prompt, then become audio through text-to-speech.Users can specify voice, theme, style, and free-form preferences.Source: §3.3 · p. 3
4 Demonstration
The demonstration lets attendees experience EarStreAM through a biosignal-adaptive mode and a personalized meditation-only mode. Its accelerated adaptive cycle uses shortened measurements and segments, while physiological processing stays on the smartphone and service prompts exclude personal and physiological information.
4 Demonstration
- Attendees use OpenEarable 2.0 and a companion smartphone to experience continuous in-ear monitoring and adaptive audio feedback.Source: §4 · pp. 3–4, Figure 3 · p. 3
- The adaptive demo shortens baseline measurement to 30 s, sets stress detection to 3%, and uses meditation segments of approximately 30 s.Optional cold-water exposure and timed mental arithmetic can induce stress; the other mode provides personalized meditation without biosignal adaptation.Source: §4 · p. 4, Figure 4 · p. 4
- Physiological processing remains local to the smartphone, and prompts sent to LLM and TTS services exclude personal and physiological information.Source: §4 · p. 4
5 Conclusion
EarStreAM combines in-ear physiological sensing, personalized LLM-generated guidance, and audio feedback in a self-contained earable system for workplace stress-adaptive meditation. The demo illustrates the system’s potential to support future closed-loop earable interventions and context-aware stress coping technologies.
5 Conclusion
- EarStreAM combines in-ear physiological sensing, personalized LLM-generated guidance, and audio feedback on OpenEarable 2.0 for workplace stress-adaptive meditation.The system is described as self-contained and intended for everyday workplace environments.Source: §5 · p. 4
- The interactive demo lets attendees experience continuous in-ear sensing and real-time adaptation for personalized stress interventions in demanding contexts.Source: §5 · p. 4
- The authors hope the demonstration motivates research on closed-loop earable interventions and context-aware technologies for timely, individualized stress support.Source: §5 · p. 4