Our study suggests an original AI-driven method for emotion recognition to improve human-computer interaction. Communication, decision-making, and behavior-shaping all depend on emotions. To accurately predict emotions across various datasets, we construct a deep learning model that examines language patterns, voice intonation, and facial expressions. Emotion recognition is critical to human-computer interaction, enabling machines to perceive and respond to users' emotional states. This comparison study evaluates various AI-driven emotion recognition techniques, including deep learning, computer vision, and natural language processing approaches. We analyze each technique's strengths, limitations, and performance metrics across multiple datasets and real-world applications. The study aims to identify the most effective and versatile emotion recognition method to enhance human-computer interaction in diverse domains.