In our digitally connected yet often isolating world, people frequently turn to online content for comfort, inspiration, and connection. Yet, the content we encounter isn’t always aligned with what we truly need in the moment. This project introduces an emotion-aware content recommendation system that to support user’s emotional well-being by offering them content that resonates with how they’re feeling. By analyzing both facial expressions and voice tones, our system senses users’ emotions using a GAN for image-based emotion detection and a Machine Learning model for audio cues. With these insights, the system fetches relevant content via the online sources, filtered to match both the detected emotions and each user’s unique preferences and interests gathered during their first interaction with the interface. This approach makes the digital experience more personal by offering content that can lift users’ spirits, calm them down, or keep them engaged, based on how they feel. By understanding each person’s emotions, the system brings a sense of care and connection to technology, making users feel truly understood and supported.