International Journal of Advances in Engineering & Scientific Research

International Journal of Advances in Engineering & Scientific Research

Print ISSN : 2349 –4824

Online ISSN : 2349 –3607

Frequency : Continuous

Current Issue : Volume 12 , Issue 2
2025

MULTIMODAL VOICE-BASED HEALTH MONITORING SYSTEM USING DEEP LEARNING

A.Wasim Raja, Bharathwaj P R, Magilavan M, Akash M , Adithya N R, Adithya J, Jeevadharshini

A.Wasim Raja, Assistant Professor, Artificial Intelligence and Data Science, Sri Krishna College of Engineering and Technology, Coimbatore, Tamilnadu Email: wasimrajaa@skcet.ac.in
Bharathwaj P R, Artificial Intelligence and Data Science, Sri Krishna College of Engineering and Technology , Coimbatore, Tamilnadu Email: bharathwajrm@gmail.com
Magilavan M, Artificial Intelligence and Data Science, Sri Krishna College of Engineering and Technology , Coimbatore, Tamilnadu Email: 727723euai063@skcet.ac.in
Akash M, Artificial Intelligence and Data Science, Sri Krishna College of Engineering and Technology ,
Coimbatore, Tamilnadu Email: 727723euai007@skcet.ac.in
Adithya N R, Artificial Intelligence and Data Science, Sri Krishna College of Engineering and Technology , Coimbatore, Tamilnadu, Email: 727723euai005@skcet.ac.in
Adithya J, Artificial Intelligence and Data Science, Sri Krishna College of Engineering and Technology , Coimbatore, Tamilnadu Email: 727723euai004@skcet.ac.in
Jeevadharshini, Artificial Intelligence and Data Science, Sri Krishna College of Engineering and Technology , Coimbatore, Tamilnadu Email: 727723euai044@skcet.ac.in

DOI : https://doi.org/10.5281/zenodo.17552230 Page No : 47-53

Published Online : 2025-11-07

Download Full Article : PDF Check for Updates


ABSTRACT

The rapid evolution of artificial intelligence and deep learning has significantly transformed the landscape of healthcare, providing novel methods for diagnosis, monitoring, and personalized treatment. Among these, voice-based health monitoring has emerged as a promising, non-invasive approach capable of providing continuous assessment of both physiological and psychological states. Human speech contains intricate patterns influenced by cardiovascular, respiratory, neurological, and

emotional conditions. This paper proposes a Multimodal Voice-Based Health Monitoring System that integrates voice signals, cough and breathing patterns, and physiological sensor data to enable real-time, accurate health monitoring.

By leveraging deep learning architectures such as Convolutional Neural Networks (CNNs), Recurrent Neural Networks (RNNs), and attention-based fusion mechanisms, the system can detect abnormalities in cardiovascular, respiratory, and mental health domains. The incorporation of multimodal data improves robustness against noise, inter-speaker variability, and environmental interference. Extensive experiments and performance evaluations demonstrate that the system achieves high accuracy and reliability, making it a viable solution for remote health monitoring, early diagnosis, and proactive healthcare interventions.

The proposed framework also considers privacy-preserving techniques and deployment strategies suitable for both cloud and edge computing environments, paving the way for scalable and secure implementation in real-world scenarios. This research contributes to bridging the gap between conventional clinical diagnostics and modern AI-powered health monitoring systems.

 

Keywords: Convolutional Neural Networks (CNNs), Recurrent Neural Networks (RNNs), Voice-Based Health Monitoring