{"ID":2869605,"CreatedAt":"2026-06-01T04:54:23.091178241Z","UpdatedAt":"2026-06-01T04:54:23.091178241Z","DeletedAt":null,"paper_url":"https://arxiv.org/abs/2509.15473","arxiv_id":"2509.15473","title":"Breathing and Semantic Pause Detection and Exertion-Level Classification in Post-Exercise Speech","abstract":"Post-exercise speech contains rich physiological and linguistic cues, often marked by semantic pauses, breathing pauses, and combined breathing-semantic pauses. Detecting these events enables assessment of recovery rate, lung function, and exertion-related abnormalities. However, existing works on identifying and distinguishing different types of pauses in this context are limited. In this work, building on a recently released dataset with synchronized audio and respiration signals, we provide systematic annotations of pause types. Using these annotations, we systematically conduct exploratory breathing and semantic pause detection and exertion-level classification across deep learning models (GRU, 1D CNN-LSTM, AlexNet, VGG16), acoustic features (MFCC, MFB), and layer-stratified Wav2Vec2 representations. We evaluate three setups-single feature, feature fusion, and a two-stage detection-classification cascade-under both classification and regression formulations. Results show per-type detection accuracy up to 89$\\%$ for semantic, 55$\\%$ for breathing, 86$\\%$ for combined pauses, and 73$\\%$overall, while exertion-level classification achieves 90.5$\\%$ accuracy, outperformin prior work.","short_abstract":"Post-exercise speech contains rich physiological and linguistic cues, often marked by semantic pauses, breathing pauses, and combined breathing-semantic pauses. Detecting these events enables assessment of recovery rate, lung function, and exertion-related abnormalities. However, existing works on identifying and disti...","url_abs":"https://arxiv.org/abs/2509.15473","url_pdf":"https://arxiv.org/pdf/2509.15473v1","authors":"[\"Yuyu Wang\",\"Wuyue Xia\",\"Huaxiu Yao\",\"Jingping Nie\"]","published":"2025-09-18T22:39:34Z","proceeding":"eess.AS","tasks":"[\"eess.AS\",\"cs.CL\",\"cs.LG\",\"cs.SD\"]","methods":"[\"LoRA\",\"Convolutional Neural Network\"]","has_code":false}