International Research Journal of Engineering and Technology (IRJET)
e-ISSN: 2395-0056
Volume: 12 Issue: 12 | Dec 2025
p-ISSN: 2395-0072
www.irjet.net
Neuro Well: A Generative AI–Powered Multimodal EmotionRecognition and Mental Health Companion System. Mr.Anirudha kolpyakwar1, Arib Rais Khan2, Nambala Murli Sai3 ,Shrutika Kiran Dusane4 , Nikita Balasaheb Matsagar5 1Mr. Anirudha Kolpyakwar (CSE Department Of Sandip University, Nashik ). 2Arib Rais Khan,3Nambala Murli Sai,4Shrutika Kiran Dusane,5Nikita Balasaheb Matsagar, (Students ,CSE
Department Of Sandip University, Nashik) ---------------------------------------------------------------------***---------------------------------------------------------------------
Abstract - Mental health challenges continue to rise globally, creating an urgent need for accessible and reliable support
systems. Advances in Artificial Intelligence and Machine Learning offer opportunities to build intelligent tools that can understand and respond to human emotions. This paper presents Nero Well, a Generative AI powered multimodal emotion recognition and mental health companion system designed to assist users through empathetic and context aware interaction. The system integrates text sentiment analysis, voice tone detection, and facial emotion recognition using deep learning models to identify the user’s emotional state with greater accuracy. These multimodal inputs are fused to enhance performance and are passed to a Generative AI catboat that delivers supportive and personalized responses. The system is developed using Python, Tensor Flow, Open CV, and Natural Language Processing techniques. Experimental evaluation shows improved emotion recognition accuracy when compared to unmoral approaches, demonstrating the effectiveness of multimodal fusion. Neuro Well is intended to function as a digital mental health companion capable of offering emotional assistance, early stress detection, and continuous wellness monitoring. This work highlights the potential of AI driven systems to support mental wellbeing by providing accessible, intelligent, and emotionally aware interaction. Key Words: Generative AI, Multimodal Emotion Recognition, Deep Learning, NLP, Mental Health, Machine Learning
1. Introduction: Mental health disorders have become a critical global concern, affecting millions of individuals and creating a growing demand for accessible, affordable, and continuous emotional support. Traditional mental health services often face challenges such as limited availability, high cost, social stigma, and lack of immediate assistance. With recent advancements in Artificial Intelligence and Machine Learning, intelligent systems now have the potential to understand human emotions and offer personalized support to users in real time. Multimodal emotion Recognition, which combines text, voice, and facial cues, significantly enhances the accuracy of identifying emotional states compared to single-channel methods. This paper introduces Neuro Well, a Generative AI powered multimodal emotion recognition and mental health companion system designed to provide empathetic interaction, early stress detection, and wellness monitoring. By integrating deep learning models, Natural Language Processing, computer vision, and a generative AI catboat, Neuro Well aims to bridge the gap between users and accessible mental health assistance. The system’s goal is to create an intelligent digital companion capable of analyzing emotional cues, understanding user context, and delivering supportive, context aware responses to promote mental well-being.
2. Literature Review: Various studies have been conducted on AI-based mental health support systems and multimodal emotion recognition technologies. Early systems were rule-based and depended heavily on predefined psychological scripts, which restricted their ability to understand complex emotional states or provide empathetic responses. With advancements in Artificial Intelligence, Natural Language Processing (NLP), and Deep Learning, modern mental health chat bots can now interpret user intent, analyze emotional cues, and adapt responses based on user interactions. Multimodal emotion recognition, which combines text, voice, and facial expressions, has gained significant importance due to its ability to capture richer emotional context compared to traditional single-modality approaches. Research by scholars such as Paul Ekman on facial expressions and recent studies in affective computing highlight the potential of deep learning models in understanding human emotions with higher accuracy. Additionally, platforms such as Tensor Flow, Py Torch, and Open CV have made it easier to develop intelligent systems capable of processing multimodal data in real time. In today’s world, individuals face increasing levels of stress, anxiety, and emotional challenges, making it difficult to seek professional help due to stigma, lack of access, and high costs. Many people struggle to find immediate
© 2025, IRJET
|
Impact Factor value: 8.315
|
ISO 9001:2008 Certified Journal
|
Page 218