Annual Conference of KIPS (한국정보처리학회:학술대회논문집)

  • 2025.05a
  • /
  • Pages.719-722
  • /
  • 2025
  • /
  • 2005-0011(pISSN)
  • /
  • 2671-7298(eISSN)
Korea Information Processing Society (한국정보처리학회)
권호 표지
DOI QR코드

DOI QR Code

TalkThru: An AI-Powered Mental Health Companion for Real-Time Emotional Support

  • Anjitha Divakaran (Dept. of AI Convergence, Chonnam National University) ;
  • Hyung-Jeong Yang (Dept. of AI Convergence, Chonnam National University) ;
  • Seungwon Kim (Dept. of AI Convergence, Chonnam National University) ;
  • Ji-eun Shin (Dept. of Psychology, Chonnam National University) ;
  • Soo-Hyung Kim (Dept. of AI Convergence, Chonnam National University)
  • Published : 2025.05.29
Download PDF
⟨ Previous Next ⟩

Abstract

Emotional connection is essential in mental health support, especially in text-based applications where users seek empathy and validation. TalkThru is an AI-powered mobile app designed to provide real-time emotional assistance through personalized conversations and self-care tools. By combining DistilBERT for emotion detection with T5 for empathetic response generation, the system ensures context-aware interaction that resonates with user emotions. A dynamic avatar, which evolves with the user’s emotional journey, adds visual engagement. The app also includes journaling, mood tracking, meditation, and music therapy, supported by gamification to encourage consistency. Developed using Kotlin and Flask, TalkThru maintains performance and scalability while securing sensitive emotional data. Initial evaluations show promising accuracy in emotion recognition and user satisfaction, positioning TalkThru as a practical, AI-driven solution for digital mental health care.

Keywords

Acknowledgement

This work was supported by Institute of Information & communications Technology Planning & Evaluation (IITP) under the Artificial Intelligence Convergence Innovation Human Resources Development (IITP-2023-RS-2023-00256629) grant funded by the Korea government (MSIT). This research was supported by the MSIT (Ministry of Science and ICT), Korea, under the ITRC (Information Technology Research Center) support program (IITP-2025-RS-2024-00437718) supervised by the IITP (Institute for Information & Communications Technology Planning & Evaluation). This work was supported by the National Research Foundation of Korea (NRF) grant funded by the Korea government (MSIT) (RS-2023-00219107).

References

  1. Henderson, C., Evans-Lacko, S., & Thornicroft, G. (2013). Mental illness stigma, help seeking, and public health programs. American journal of public health, 103(5), 777-780.
  2. Fitzpatrick, K. K., Darcy, A., & Vierhile, M. (2017). Delivering cognitive behavior therapy to young adults with symptoms of depression and anxiety using a fully automated conversational agent (Woebot): a randomized controlled trial. JMIR mental health, 4(2), e7785.
  3. Inkster, B., Sarda, S., & Subramanian, V. (2018). An empathy-driven, conversational artificial intelligence agent (Wysa) for digital mental well-being: real-world data evaluation mixed-methods study. JMIR mHealth and uHealth, 6(11), e12106.
  4. Hannah Rashkin, Eric Michael Smith, Margaret Li, and Y-Lan Boureau. 2019. Towards Empathetic Open-domain Conversation Models: A New Benchmark and Dataset. In Proceedings of the 57th Annual Meeting of the Association for Computational Linguistics, pages 5370–5381, Florence, Italy. Association for Computational Linguistics.
  5. Sharma, A., Lin, I. W., Miner, A. S., Atkins, D. C., & Althoff, T. (2021, April). Towards facilitating empathic conversations in online mental health support: A reinforcement learning approach. In Proceedings of the web conference 2021 (pp. 194-205).
  6. Kshitij Mishra, Priyanshu Priya, and Asif Ekbal. 2023. PAL to Lend a Helping Hand: Towards Building an Emotion Adaptive Polite and Empathetic Counseling Conversational Agent. In Proceedings of the 61st Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers), pages 12254–12271, Toronto, Canada. Association for Computational Linguistics.
  7. O'neil, E., Sedoc, J., Yang, D., Zhu, H., & Ungar, L. (2023, December). Automatic Reflection Generation for Peer-to-Peer Counseling. In Proceedings of the Third Workshop on Natural Language Generation, Evaluation, and Metrics (GEM) (pp. 62-75).
  8. Mishra, K., Priya, P., & Ekbal, A. (2023, June). Help me heal: A reinforced polite and empathetic mental health and legal counseling dialogue system for crime victims. In Proceedings of the AAAI Conference on Artificial Intelligence (Vol. 37, No. 12, pp. 14408-14416).
  9. Priya, P., Mishra, K., Totala, P., & Ekbal, A. (2023, January). PARTNER: A Persuasive Mental Health and Legal Counselling Dialogue System for Women and Children Crime Victims. In IJCAI (pp. 6183-6191).
  10. Sanh, V., Debut, L., Chaumond, J., & Wolf, T. (2019). DistilBERT, a distilled version of BERT: smaller, faster, cheaper and lighter. ArXiv, abs/1910.01108.
  11. Colin Raffel, Noam Shazeer, Adam Roberts, Katherine Lee, Sharan Narang, Michael Matena, Yanqi Zhou, Wei Li, and Peter J. Liu. 2020. Exploring the limits of transfer learning with a unified text-to-text transformer. J. Mach. Learn. Res. 21, 1, Article 140 (January 2020), 67 pages.