Recognition of energy and/or sound related activities performed by aged

In this final year project, the author explored the application of deep learning for audio classification, with a focus on multi-class and multi-label classification of various sound classes found in audio files. Examples of these sound classes include door sounds, toilet flushing, washing machines, and other typical household noises. The primary objective was to monitor the daily activities of elderly individuals by analyzing audio captured from their electronic devices, offering valuable insights into their daily routines, and helping to detect potential safety concerns. The author employed convolutional neural networks (CNNs) in the project, as they are highly effective in image and signal processing tasks, making them suitable for audio classification problems. The project also involved experimenting with data augmentation and pre-processing techniques to enhance the model's performance and resilience to different noises and disturbances. To develop the deep learning model, the author collected several labelled audio samples from a variety of sources and divided them into training, validation, and testing sets, ensuring a fair evaluation of the model's performance. Metrics such as accuracy, F1-score, and confusion matrix were utilized to assess the model's efficiency. The results suggested that the proposed approach can effectively detect ambient sounds, forming the basis for monitoring elderly activities and identifying unusual patterns, potentially contributing to their safety and well-being. In conclusion, this final year project demonstrates the potential of deep learning techniques, particularly CNNs, for audio classification in monitoring the daily activities of elderly individuals using audio data from their mobile/electronic devices. The outcomes of this project hold significant implications for elderly care support and offer valuable information for researchers and healthcare professionals in the field of geriatrics.Bachelor of Engineering (Electrical and Electronic Engineering