Abstract
<jats:p>Popular music reflects and shapes cultural dynamics, yet research on trends in music over long periods has focused almost exclusively on lyrics. We analyze diachronic trends in eight Spotify audio features (danceability, energy, loudness, speechiness, acousticness, instrumentalness, liveness, and valence) for the most popular songs in the United States (Billboard Hot 100) and Germany (GfK Entertainment Charts) from 1977 to 2025. Both markets exhibit broadly parallel trajectories: a steady decline in valence, an increase in loudness followed by a reversal in the streaming era, and a pattern of decline and subsequent recovery in acousticness. A decomposition at the level of genres reveals that the reversal in loudness and the decline in valence operate within genres rather than reflecting compositional shifts. We find that synchronization of audio features across the two countries, measured through rolling correlations of weekly series in first differences, has increased substantially since the middle of the 2010s, coinciding with the dominance of digital streaming platforms. This synchronization is strongest for features related to production and weakest for features that depend on language. Decomposing the effect, we show that the increase is driven both by a growing overlap in charted repertoire and by a genuine acoustic convergence among songs exclusive to one country, suggesting that streaming platforms synchronize national music markets through both a direct repertoire channel and an indirect aesthetic channel.</jats:p>