This study presents a computational approach to generating playlist cover images by leveraging lyric-based semantic analysis. We employ natural language processing techniques, including part-of-speech tagging, lemmatization, and K-means clustering, to extract thematic and emotional features from song lyrics. These extracted features are then transformed into structured prompts for text-to-image models, enabling the automated generation of visually representative playlist cover art. To evaluate the effectiveness of the generated images, we conducted a user study that assessed aesthetic quality, thematic coherence, and user satisfaction. Our results indicate that AI-driven lyric analysis can produce compelling visual representations aligned with musical identity. This framework has potential applications in music streaming platforms, digital media curation, and personalized content generation, offering a novel intersection between computational creativity and AI-driven design.