whatImeant.: Designing an AI-powered Interface for Music Semantics Discovery

Since ancient times, music has been a faithful companion to people, capable of inspiring, accompanying, and comforting. While listening to music was once restricted by the limited resources of the past, today it has become so accessible that it is often taken for granted. Although this high accessibility has made music an almost constant presence in our daily lives, it has also reduced the level of attention paid to it. Many people fall in love with the melody before the lyrics, and most of them never actually explore the semantic meaning of the song. In a world where free time is increasingly limited and attention spans are getting shorter, even a simple action like looking up the meaning of a specific word, verse, or song feels like a tedious and heavy task, even with the web's infinite database of information readily available. As a result, the desire for discovery and knowledge fades away, turning into an occasional event. The goal of whatImeant. is to make researching and understanding songs (and the artists behind them) extremely convenient and immediate, removing the barrier of laziness that comes from having to manually search for information. It is designed as a second-screen application that enhances the streaming platform experience by combining real-time playback data extraction via APIs with semantic analysis capabilities generated by LLMs (Large Language Models). While the backend infrastructure handles lyric retrieval and AI integration, this paper focuses on the architecture and development of the Frontend component. This work covers the development process of a modern and reactive user interface, capable of transforming complex data streams and asynchronous processing times into a smooth, intuitive, and uninterrupted visual experience for the listener.