Lightweight Audio-Based Human Activity Classification Using Transfer Learning

Marco Nicolini, Federico Simonetta, Stavros Ntalampiras

2023

Abstract

This paper employs the acoustic modality to address the human activity recognition (HAR) problem. The cornerstone of the proposed solution is the YAMNet deep neural network, the embeddings of which comprise the input to a fully-connected linear layer trained for HAR. Importantly, the dataset is publicly available and includes the following human activities: preparing coffee, frying egg, no activity, showering, using microwave, washing dishes, washing hands, and washing teeth. The specific set of activities is representative of a standard home environment facilitating a wide range of applications. The performance offered by the proposed transfer learning-based framework surpasses the state of the art, while being able to be executed on mobile devices, such as smartphones, tablets, etc. In fact, the obtained model has been exported and thoroughly tested for real-time HAR on a smartphone device with the input being the audio captured from its microphone.

Paper Citation

in Harvard Style

Nicolini M., Simonetta F. and Ntalampiras S. (2023). Lightweight Audio-Based Human Activity Classification Using Transfer Learning. In Proceedings of the 12th International Conference on Pattern Recognition Applications and Methods - Volume 1: ICPRAM, ISBN 978-989-758-626-2, pages 783-789. DOI: 10.5220/0011647900003411

in Bibtex Style

@conference{icpram23,
author={Marco Nicolini and Federico Simonetta and Stavros Ntalampiras},
title={Lightweight Audio-Based Human Activity Classification Using Transfer Learning},
booktitle={Proceedings of the 12th International Conference on Pattern Recognition Applications and Methods - Volume 1: ICPRAM,},
year={2023},
pages={783-789},
publisher={SciTePress},
organization={INSTICC},
doi={10.5220/0011647900003411},
isbn={978-989-758-626-2},
}

in EndNote Style

TY - CONF

JO - Proceedings of the 12th International Conference on Pattern Recognition Applications and Methods - Volume 1: ICPRAM,
TI - Lightweight Audio-Based Human Activity Classification Using Transfer Learning
SN - 978-989-758-626-2
AU - Nicolini M.
AU - Simonetta F.
AU - Ntalampiras S.
PY - 2023
SP - 783
EP - 789
DO - 10.5220/0011647900003411