A Review Paper on Dual-Mode Emotion Recognition Systems Using Facial Analysis and Interactive Questioning

Emotion recognition has become a significant research area in computer vision and affective computing due to its growing applications in human computer interaction. Early approaches mainly relied on facial expression analysis; however, recent studies emphasize multimodal and contextual information for improved robustness. The review paper analyses recent advancements in emotion recognition systems with a focus on dual mode emotion approaches combining facial and speech-based emotion recognition. The study reviews deep learning techniques, system architectures, and commonly used datasets, particularly RAVDESS, for speech emotion recognition. It discusses an implemented facial expression recognition module to connect theory with practical use. The paper highlights existing research gaps concerning real-time performance, robustness, and computational efficiency. Finally, it outlines future research directions, focusing on efficient multimodal fusion, improved accuracy, and systems that can recognize emotion in real time.