Silent, eyes-free text entry remains challenging when speech and conventional touch input are impractical. Prior wearable systems often required custom sensors or limited users to a small vocabulary. We present MorsEar, an IMU-only earable framework that maps near-ear micro-gestures such as taps for dot/dash; slide/pull/circle for space/delete/send into character-level Morse, enabling unrestricted character composition while using a compact lexicon solely for lightweight on-device autocorrect. The result is a low-bandwidth, reduced-exposure communication channel that works eyes-free and voice-free in accessibility scenarios, silent zones, and constrained environments. MorsEar infers words using a physics-aware preprocessing stack and compact CNN feed a tempo-adaptive segmentation with rolling buffers; an on-device decoder with lightweight autocorrect provides real-time feedback entirely on-phone. In a 24-participant study (with four accessibility users) across Silent, Cafe, and Metro, MorsEar achieved CER 7.3% and WER 12.5% → 7.8% (Autocorrect), with median 9.3/9.1/5.8 WPM, respectively. Similar to other accessibility-oriented encodings such as Braille, Morse requires a brief familiarization period to learn the timing and rhythm of dots and dashes; after which, MorsEar shows that commodity earable IMUs can support discreet, low-exposure text entry that scales beyond discrete commands to language-level interaction.

Federated Learning (FL) enables collaborative machine learning across decentralized devices and data sources, but resource constraints on pervasive devices necessitate efficient model compression. Existing approaches, such as quantization for on-device training, often degrade accuracy, especially for classes that are difficult to learn due to imbalance, poor-quality samples, or inherent complexity. This results in persistent accuracy gaps across classes. We propose Fed-CASQ's a novel framework that couples class-aware strategies into the quantization process to jointly improve efficiency and accuracy in pervasive FL. Unlike prior works that address quantization and imbalance separately, Fed-CASQ adaptively selects quantization levels based on device resources and leverages Layer-wise Relevance Propagation (LRP) to assess class-relevant convolutional neural network (CNN) filters on the client side. An adaptive weight scaling mechanism is then applied to amplify critical information for low-accuracy classes before aggregation. At the server, a complementary novel aggregation strategy mitigates global imbalance across clients, ensuring that underperforming classes receive proportional attention during model updates. We theoretically establish that Fed-CASQ achieves a convergence rate of ${\mathcal{O}}\left({\frac{{\kappa *\hat \sigma *\hat \delta }}{{\sqrt T }}}\right)$ under non-convex settings. We empirically establish that quantization directly influences the performance of under sampled (minority) classes. Experimental results further show that Fed-CASQ substantially narrows the performance gap for low-accuracy classes, improving their accuracy by ≈30%, while reducing training latency by over 56% on resource-constrained pervasive devices.

Wrist micro-behaviors during daily activities such as typing, handwriting, cooking, or carrying objects are valuable indicators for early detection of wrist disorders like Carpal Tunnel Syndrome and tendonitis. However, continuous personalized monitoring remains challenging without intrusive setups or hand-crafted rules. We present WristSense, a real-time, wrist-worn sensing system that introduces: (i) a magnetometer-stabilized quaternion fusion pipeline for orientation-agnostic tracking, (ii) a lightweight 1D-CNN + HMM model to distinguish functional gestures from strain-related coping behaviors, and (iii) an inertial tokenization scheme that converts events into structured prompts for a pretrained LLM. This enables zero-shot ergonomic feedback without per-user calibration. Evaluations across 12 participants show accurate posture tracking (< 10° MAE), high gesture recognition (macro F1 = 0.91), and improved usability (SUS = 85.2), with significantly higher user compliance compared to rule-based methods. WristSense demonstrates the potential of combining inertial sensing with LLMs for scalable, personalized ergonomic monitoring and early intervention.

We present SpineSense , an interactive earable-based framework for continuous monitoring of cervical spine posture and discomfort-related behavior in daily life. Leveraging inertial data from commercial off-the-shelf (COTS) earables (e.g., Apple AirPods Pro 2), SpineSense models the cervical vertebral chain (C1–C7) using SLERP-interpolated quaternions to estimate craniovertebral (CV) angles and identify early pain-relief gestures (e.g., neck rubbing, circular head rolls) that are associated with early signs of musculoskeletal fatigue, as informed by clinical observation. A real-time feedback loop delivers posture-based alerts and weekly compliance summaries to support user awareness and long-term posture correction. Central to our design is a clinician-in-the-loop methodology: clinical experts (including orthopedic surgeons and physiotherapists) informed threshold selection (e.g., CV angle cutoffs), interpreted common discomfort behaviors, and iteratively guided the structure of weekly feedback reports. We evaluate the system on 20 participants and achieve low spine angle estimation error (MAE: 0.876° , RMSE: 1.02° , r = 0.95), and discomfort gesture classification accuracy of F 1 = 0.97. Usability studies across diverse activities show high acceptance (SUS = 84.75, NASA-TLX = 32.7, PSSUQ = 2.13), with formal ANOVA tests validating statistically significant improvements over baseline interfaces. Together, our findings establish the feasibility of engineering interactive cervical health systems using COTS earables that support real-time feedback, clinician-informed reporting, and pervasive deployment in naturalistic settings.

Developing scalable wearable human activity recognition (wHAR) models is challenging due to domain shifts that substantially degrade performance across downstream tasks. Unsupervised domain adaptation (UDA) seeks to improve generalization by transferring knowledge from labeled source domains to unlabeled target domains. However, conventional UDA methods primarily align marginal feature distributions while neglecting feature-label dependencies, often leading to negative transfer and sub-optimal performance. Motivated by these limitations, we propose a novel optimization framework that tackles two key challenges: (i) generating reliable pseudo-labels for the unlabeled target domain and (ii) minimizing conditional discrepancies across domains. To address (i), we employ temperature-based entropy minimization (TEM), which calibrates prediction confidence by scaling logits with a temperature parameter to produce robust pseudo-labels. For (ii), we introduce a polynomial kernel-based cross-covariance (PkCC) loss, a high-order statistics-driven approach that maps features into a reproducing kernel hilbert space (RKHS) to capture richer feature-label dependencies and reduce conditional distribution gaps between domains. In addition, we demonstrate that CoDAN readily extends to partial UDA (pUDA), where the target label space is a subset of the source, and extensive evaluations on public wHAR datasets with diverse label spaces validate its superior performance over state-of-the-art methods in both UDA and pUDA scenarios.

Automated dietary monitoring is essential for gaining insights into eating behaviors, especially for managing chronic conditions such as obesity, diabetes, and hypercholesterolemia. Earable-based inertial sensing has been found promising for detecting chewing and eating activities; however, further insights like what, when, and how much is being eaten are crucial information for effective dietary assessment. Therefore, we propose BiteSense, an earable-based system that leverages inertial sensors (IMU) to monitor food intake and classify various food types. Using a hierarchical classification model, the system analyzes masticatory kinematics to detect food states, textures, nutritional value, and cooking methods, ultimately identifying specific foods consumed, as well as estimating food intake amount and meal type. A semi-controlled user study involving 38 participants from diverse backgrounds demonstrated the system’s high accuracy, with an F1 score of 0.86 for detecting the masticatory process using a leave-one-subject-out (LOSO) approach, while exhibiting significant improvement over benchmark algorithms in extensive experiments by 8-12%.

Recent advancements in deep learning-based wearable human action recognition (wHAR) have improved the capture and classification of complex motions, but adoption remains limited due to the lack of expert annotations and domain discrepancies from user variations. Limited annotations hinder the model's ability to generalize to out-of-distribution samples. While data augmentation can improve generalizability, unsupervised augmentation techniques must be applied carefully to avoid introducing noise. Unsupervised domain adaptation (UDA) addresses domain discrepancies by aligning conditional distributions with labeled target samples, but vanilla pseudo-labeling can lead to error propagation. To address these challenges, we propose μDAR, a novel joint optimization architecture comprised of three functions: (i) consistency regularizer between augmented samples to improve model classification generalizability, (ii) temporal ensemble for robust pseudo-label generation and (iii) conditional distribution alignment to improve domain generalizability. The temporal ensemble works by aggregating predictions from past epochs to smooth out noisy pseudo-label predictions, which are then used in the conditional distribution alignment module to minimize kernel-based class-wise conditional maximum mean discrepancy (kCMMD) between the source and target feature space to learn a domain invariant embedding. The consistency-regularized augmentations ensure that multiple augmentations of the same sample share the same labels; this results in (a) strong generalization with limited source domain samples and (b) consistent pseudo-label generation in target samples. The novel integration of these three modules in μDAR results in a range of ~ 4-12% average macro-F1 score improvement over six state-of-the-art UDA methods in four benchmark wHAR datasets.