Real time speech emotion recognition for Bio-medical monitoring using hybrid deep learning architecture
International Journal of Computational and Experimental Science and Engineering · 30.11.2025 · 11 (4)
DOI: 10.22399/ijcesen.5205 ↗Kurz gefasst
Fachartikel in der Zeitschrift International Journal of Computational and Experimental Science and Engineering, online erschienen am 30.11.2025. Verfasst von Dhaval R Bhojani und 4 weiteren Autor:innen. Die Arbeit nennt 33 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0).
Abstract (Originalsprache)
Speech carries essential emotional information that is significant for biomedical monitoring and treatment of patients. Real-time SER is now a vital medical tool, allowing simultaneous evaluation of psychological states. The anticipated approach in this research is a DL model that integrates CNNs and LSTMs to enhance the accuracy of emotion recognition by extracting spectral and temporal features of speech signals. Most conventional SER models are plagued by low accuracy, poor noise adaptability, and poor temporal modeling, thus constraining their usability in biomedical situations. The proposed "CNN–LSTM model" overcomes such challenges by utilizing the efficient feature extraction ability of CNN and sequence model ability of LSTM to offer strong and real-time classification of emotional states. The model was trained and tested on the "RAVDESS dataset" with data augmentation and preprocessing for better generalization. Experimental results confirm that the model is 92.8% accurate, 91.6% precise, 92.1% recalled, and 91.8% F1-scored, outperforming existing solutions. This attests that it is well-suited to be implemented in healthcare platforms for real-time biomedical monitoring, stress detection, and adaptive interventions.
Zitieren
APA
Bhojani, D. R., J.Trivedi, R., Sangani, D. J., Jani, D. G., & Gohil, N. B. (2025). Real time speech emotion recognition for Bio-medical monitoring using hybrid deep learning architecture. International Journal of Computational and Experimental Science and Engineering, 11(4). https://doi.org/10.22399/ijcesen.5205
Vancouver
Bhojani DR, J.Trivedi R, Sangani DJ, Jani DG, Gohil NB. Real time speech emotion recognition for Bio-medical monitoring using hybrid deep learning architecture. International Journal of Computational and Experimental Science and Engineering. 2025;11(4). doi:10.22399/ijcesen.5205
BibTeX
@article{Bhojani2025,
author = {Bhojani, Dhaval R and J.Trivedi, Rushi and Sangani, Dhara J. and Jani, Devang G and Gohil, N B},
title = {Real time speech emotion recognition for Bio-medical monitoring using hybrid deep learning architecture},
journal = {International Journal of Computational and Experimental Science and Engineering},
year = {2025},
volume = {11},
number = {4},
doi = {10.22399/ijcesen.5205}
}
EUVIMED