发表机构
National Institute of Technology Durgapur（杜尔加布尔国立理工学院）
机构由 AI 辅助整理，请以论文原文为准。AI 中文总结
针对辅助生活的浴室声学事件识别，提出紧凑可解释的原始波形分类器SincDPNet及七类数据集，经贝叶斯优化权衡性能与规模，最佳模型达80.2%准确率。
AI 中文摘要
浴室声学事件识别可以在不希望进行连续视频监控的环境中支持环境辅助生活。然而，实际部署需要模型具备紧凑性、可解释性以及对录音环境变化的鲁棒性。本工作引入了\dataset{}，一个包含21,387个标注片段、在五个环境中录制的七类浴室声学事件数据集，并提出了SincDPNet，一种紧凑的原始波形分类器，由可学习的sinc滤波器组和深度可分离卷积主体组成。每个sinc滤波器由两个频率参数控制，使得学习到的通带可以直接以赫兹为单位检查，同时保持前端较小。为了减少特定房间的泄漏，在将重叠窗口分配到训练、验证和测试分区之前，先对录音会话和环境进行分离。我们进一步使用多目标贝叶斯优化作为设计工具，检查24种配置下的验证性能与模型大小之间的权衡。选定的设计覆盖了不同的工作点：性能最佳的模型在14,040个参数下达到80.2%的准确率和0.760的宏F1分数，而紧凑的$N_f=25$配置仅使用2,848个参数，在保留环境上达到75.7%的准确率、0.661的宏F1分数和0.716的MCC。对学习到的滤波器和混淆模式的分析表明，频谱重叠导致水相关事件之间的混淆，而\textit{Door}/\textit{Walker/Crutch}错误也反映了它们瞬态时间结构的相似性。
英文摘要
Bathroom acoustic-event recognition can support ambient assisted living in settings where continuous video monitoring is undesirable. However, practical deployment requires models that are compact, interpretable, and robust to changes in the recording environment. This work introduces \dataset{}, a seven-class bathroom acoustic-event dataset containing 21{,}387 annotated clips recorded across five environments, and proposes SincDPNet, a compact raw-waveform classifier with a learnable sinc filter bank followed by a depthwise-separable convolutional body. Each sinc filter is controlled by two frequency parameters, allowing the learned passbands to be inspected directly in hertz while keeping the front end small. To reduce room-specific leakage, recording sessions and environments are separated before overlapping windows are assigned to the training, validation, and test partitions. We further use multi-objective Bayesian optimization as a design tool to examine the validation performance--model-size trade-off across 24 configurations. The selected designs span different operating points: the best-performing model achieves 80.2\% accuracy and 0.760 macro-F1 with 14{,}040 parameters, while the compact $N_f=25$ configuration uses only 2{,}848 parameters and achieves 75.7\% accuracy, 0.661 macro-F1, and 0.716 MCC on the held-out environment. Analysis of the learned filters and confusion patterns shows that spectral overlap contributes to confusion among water-related events, while the \textit{Door}/\textit{Walker/Crutch} errors also reflect similarities in their transient temporal structure.
Comments29 pages, 26 figures
