EarResp-ANS : Audio-Based On-Device Respiration Rate Estimation on Earphones with Adaptive Noise Suppression

2026-02-03

AI Summary

This work proposes a seamless, low-power, on-device algorithm for respiratory rate estimation using in-ear headphone audio signals, addressing the limitations of existing methods that often suffer from high power consumption, privacy concerns, or inadequate performance in real-world noisy environments. The system operates entirely on commercial off-the-shelf headphones without transmitting audio data externally or relying on deep learning models. By integrating LMS-based adaptive noise cancellation, physiological acoustic feature extraction, and lightweight signal analysis, it achieves robust performance even under environmental noise levels up to 80 dB SPL. In experiments with 18 participants, the method yields a global mean absolute error (MAE) of 0.84 breaths per minute (CPM), which improves to 0.47 CPM after outlier removal, while maintaining a processor load below 2%, thereby ensuring high accuracy, energy efficiency, and strong privacy preservation.

Technology Category

• Machine Learning: Hardware-aware ML
• Planning, Routing, and Scheduling: Activity and Plan Recognition
• Intelligent Robots: State Estimation

Application Category

• User Modeling, Personalization and Recommendation: On-Device user modeling, personalization, and recommendation
• Security and Privacy: Large-scale security measurements
• Responsible Web: Algorithmic accountability and transparency on the web

Abstract

Respiratory rate (RR) is a key vital sign for clinical assessment and mental well-being, yet it is rarely monitored in everyday life due to the lack of unobtrusive sensing technologies. In-ear audio sensing is promising due to its high social acceptance and the amplification of physiological sounds caused by the occlusion effect; however, existing approaches often fail under real-world noise or rely on computationally expensive models. We present EarResp-ANS, the first system enabling fully on-device, real-time RR estimation on commercial earphones. The system employs LMS-based adaptive noise suppression (ANS) to attenuate ambient noise while preserving respiration-related acoustic components, without requiring neural networks or audio streaming, thereby explicitly addressing the energy and privacy constraints of wearable devices. We evaluate EarResp-ANS in a study with 18 participants under realistic acoustic conditions, including music, cafeteria noise, and white noise up to 80 dB SPL. EarResp-ANS achieves robust performance with a global MAE of 0.84 CPM , reduced to 0.47 CPM via automatic outlier rejection, while operating with less than 2% processor load directly on the earphone.

Problem

• Respiratory Rate Estimation
• On-Device Sensing
• Adaptive Noise Suppression
• In-Ear Audio Sensing
• Wearable Health Monitoring

Innovation

• on-device respiration estimation
• adaptive noise suppression
• earphone-based sensing
• LMS algorithm
• low-power wearable