AI Chatbot with Real Time Emotion Sensing
Vijay Kumar Shah, Jatin Sahu, Smita Gupta, Kavita Paswan, Kamini Maheshwari, Neha Lidoriya · 2025
This paper presents a novel approach to combating loneliness by introducing an emotion recognition-based personalized chatbot. The chatbot serves as a virtual companion, enabling users to express themselves freely and fostering emotional connection through realistic and empathetic interactions. By integrating deep learning techniques for emotion detection with advanced natural language processing, the platform dynamically adapts its communication style to match the user's emotional state and preferences. This research highlights the platform’s innovative features, technical workflow, and potential applications in mental health and personalized virtual companionship. The emotion detection module achieves an accuracy of 94.8% across diverse emotional states, evaluated on a custom- labeled dataset of 50,000 facial images. Integration with the chatbot enables real-time emotional adaptability, reducing response latency to 1.2 seconds. The system significantly enhances user engagement and emotional satisfaction, with surveys indicating a 38% improvement compared to standard non-emotion-adaptive chatbots. Our results demonstrate the effectiveness of coupling advanced emotion recognition with generative conversational AI, offering a transformative application in human-computer interaction.