Loading article...
Loading article...
Stream/Area:
-
Subject:
-
Paper ID:
IJSREM63646
Volume:
10 |
5 |
May 2026
Pages:
1-7
Publication Frequency:
Monthly (12 issues per year)
License:International Journal of Scientific Research in Engineering and Management follows the Creative Commons Attribution 4.0 (CC BY 4.0) License, allowing free use and distribution with proper attribution.
Bachelor of Technology
Vivekananda Global University
Bachelor of Business Administration
Vivekananda Global University
Bachelor of Technology
Vivekananda Global University
Bachelor of Technology
Vivekananda Global University
Bachelor of Technology
Vivekananda Global University
With the increasing levels of stress and psychological distress, the need for effective and easy-to-use monitoring systems is already evident. This article describes EmoSense AI, a multi-faceted system that analyzes voice and text data to detect mood and stress levels . Unlike traditional approaches that rely on physical sensors, EmoSense AI uses natural, everyday conversations. This makes it realistic, scalable, and suitable for real-world use. The system processes acoustic features extracted from speech signals, such as timbre, power, and multi-frequency cepstrum coefficients (MFCCs). In addition, it processes semantic and emotion-related features extracted from text using natural language processing (NLP) techniques. Emotion classification is performed using a deep learning framework. A long-term memory network (LSTM) processes the text input, and a hybrid convolutional neural network model (LSTM) processes the speech. In the next step, the results of both methods are combined using a “delayed fusion” technique to improve the overall reliability and accuracy of the recognition process. [2], [3] The results showed that the combined use of voice and text helped to better understand emotional states than voice or text-based methods alone. From student mental health monitoring and workplace wellness programs to intelligent conversational agents, EmoSense AI has a wide range of potential applications. In addition, it has the potential to become a proactive tool for early diagnosis of stress.
KEYWORDS
The mental health technology sector is experiencing significant growth. The rapid growth of e-commerce, increasing public awareness, the rapid spread of digital devices, and the availability of accessible mental health solutions are fueling demand for them worldwide. According to recent market research estimates, the mental health app market is expected to reach an estimated $17.5 billion by 2030, indicating continued and rapid growth in demand across diverse populations and regions. [14] Title This market is growing steadily and is estimated to grow from $5.2 billion in 2023 to $17.5 billion by 2030. The following factors are responsible for this significant development: Raising awareness about mental health issues The use of smartphones and mobile applications is increasing. There is a growing demand for solutions that allow for the delivery of telemedicine and digital healthcare. Medium-term forecast: $8.9 billion by 2025 and $13.1 billion by 2028. M-Sense Artificial Intelligence is a huge opportunity . The demand for scalable, scalable, and intelligent mental health solutions is growing rapidly worldwide. M-Sense Artificial Intelligence is uniquely positioned to meet this urgent need.
[1] S. Boria, E. Cambria, R. Bajpai, A. Hussain, “Review of sentiment analysis: From single-method analysis to multi-method fusion”, Information Fusion, vol. 1, pp. 101-37, pp. 98-125, 2017. [2] One. [3] S. Hoekreiter and J. Schmittuber , "Long-term memory", Neural Computation, Volume 9, Issue 8, pp. 1735-1780, 1997. [4] P. McAfee et al., “ Librosa : Speech and Music Signal Analysis with Python,” Proceedings of the 14th Python in Science Conference, Austin, Texas, USA, 2015, pp. 18–25. [5] A. Vaswani et al., “Attention is everything”, Advances in Neural Information Processing Systems ( Neurops ), 2017, vol. 1, p. 101. See also ba on 30, pp. 5998-6008. [6] J. Devlin, M. W. K. Ba. Chang, K. Li, K. Dotanova, “BART: Pre-training a Deep Bidirectional Transformer for Language Understanding,” Proc. NAACL-HLT, Minneapolis, Minnesota, June 2019, pp. 4171–4186. [7] R. Koe et al., “Emotion Recognition in Human-Computer Interaction”, IEEE Journal of Signal Processing, Volume 18, Issue 1, pp. 32-80, January 2001. [8] Z. Zhang, “Towards robust multi-modal sentiment analysis”, Proceedings of the ACM ICMI, Boulder, Colorado, October 2018, pp. 347-355. [9] M. Abadi et al., “TensorFlow: A Framework for Large-Scale Machine Learning”, Proceedings of the 12th USNIX OSDI Conference, Savannah, Georgia, November 2016, pp. 265-283. [10] KR Scherer, "Verbal Emotion Expression: A Review and Model for Future Research", Psychological Bulletin, Volume 99, Issue 2, pp. 143-165, 1986.
Publication Date
5/22/2026
Peer Review Type
Double-Blind Peer Review
Review Rounds: 2
Documents
Article Views
0
Total views since publication
PDF Downloads
2
Total downloads since publication
Geographic Reach
5
Digital Preservation: This article is permanently archived on Zenodo and complies with all data preservation standards across major indexing platforms.
Plagiarism Screening
This manuscript has been checked for plagiarism using TurnitinPlagiarismChecker software.
Conflict of Interest
The authors declare no conflict of interest related to this research work.
Funding Information
No funding information available for this research.
Related research in the same subject area
Loading similar articles...