EUVIMEDEUVIMED entdecken
Crossref

ICCICT 2026_AI Powered Mental Health Assessment Using Speech and Text Analysis

Banupriya C B, T.Jayasree, Dr.K.Santhi

i-manager's Journal on Digital Signal Processing · 20.06.2026 · 14 (1) · 1

DOI: 10.26634/jdp.14.1.1292 ↗

Kurz gefasst

Fachartikel in der Zeitschrift i-manager's Journal on Digital Signal Processing, online erschienen am 15.06.2026. Verfasst von Banupriya C B und 2 weiteren Autor:innen.

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

An estimated 280 million people worldwide have mental health disorders, such as anxiety or depression; however, early detection of these disorders can be challenging because of the subjective nature of traditional methods of mental health assessment (such as clinical interviews and self-report questionnaires) and their limited accessibility. This research proposes a scalable, objective, and non-invasive AI-Based mental health assessment framework that utilizes NLP and speech signal analysis (SSA) to identify the psychological condition(s) of individuals. The study utilizes natural language processing using transformer models (BERT) to identify semantic and contextual patterns in written text and uses long short-term memory (LSTM) networks to identify pitch, jitter, shimmer, and Mel-frequency cepstral coefficients (MFCCs) in speech in order to recognize and interpret emotional and psychological states. The framework is trained using a mixture of online data (Reddit posts), therapy session transcripts, and the DAIC-WOZ database, which enables the system to identify the earliest indicators of depression and anxiety. The multi-modal fusion layer integrates information from both text and speech inputs to provide the most accurate and reliable predictions for the user's mental health status. The final product of the research will be available as a real-time, mobile-friendly application that allows users to submit both voice and text data for mental health screening along with feedback, risk assessments, and personalized recommendations for mental health resources.

Zitieren

APA

B, B. C., T.Jayasree, & Dr.K.Santhi (2026). ICCICT 2026_AI Powered Mental Health Assessment Using Speech and Text Analysis. i-manager's Journal on Digital Signal Processing, 14(1), 1. https://doi.org/10.26634/jdp.14.1.1292

Vancouver

B BC, T.Jayasree, Dr.K.Santhi. ICCICT 2026_AI Powered Mental Health Assessment Using Speech and Text Analysis. i-manager's Journal on Digital Signal Processing. 2026;14(1):1. doi:10.26634/jdp.14.1.1292

BibTeX
@article{B2026,
  author = {B, Banupriya C and T.Jayasree and Dr.K.Santhi},
  title = {ICCICT 2026_AI Powered Mental Health Assessment Using Speech and Text Analysis},
  journal = {i-manager's Journal on Digital Signal Processing},
  year = {2026},
  volume = {14},
  number = {1},
  pages = {1},
  doi = {10.26634/jdp.14.1.1292}
}
Kostenlos starten →