A study analyzing all finalist songs from the 75 editions of the Sanremo Music Festival reveals a gradual move toward increasing semantic uniformity in Italian popular music lyrics.
The research employs a flexible methodology combining full-text, segment-based, topic-based, and word-level analyses using embedding techniques and large language models to track changes in semantic similarity over time.
These findings echo global patterns of declining semantic variety observed in English-language songs on streaming platforms, underscoring the value of natural language processing tools in uncovering long-term shifts in musical language.