Abstract

Musicians have long sought technology to facilitate and develop their abilities to maintain stable tempos. The rise of inherently tempo-stable instruments and performance technologies in the late 20th century—for example, turntables, sequencers, drum machines, click tracks, and post-production quantization—has fed an increasing trend toward tempo stability in Western popular music. However, the exact progression of this historical change has not been systematically documented. In this article, we report a historical analysis of tempo stability in recordings of Western Classical and Anglo-American popular music released between 1920 and 2020, in particular testing the hypothesis that tempo in certain popular genres became significantly more stable in the period 1980–2000. We distinguish between gradual changes of tempo that are not perceived as discontinuities ( tempo drift) and abrupt changes of tempo ( tempo shifts), and seek to quantify both types of tempo instability. Using the Spotify web API, we sampled measure-by-measure tempo estimates from over 45,000 tracks in seven genres: Blues, Classical, Country, Electronic Dance Music (EDM), Hip hop, Jazz, and Rock. Our analyses reveal significant changes in tempo stability in Rock, Country, and Blues music, consistent with our hypothesis. We next propose, compare, and evaluate several approaches to identifying artificially tempo-stable tracks from their tempo estimates, using the opposition between Classical and Hip hop as a faux ground truth. Data, analyses, and interactive visualizations are available for inspection on our website https://ccml.music.gatech.edu/TempoStability .

Full Text
Published version (Free)

Talk to us

Join us for a 30 min session where you can share your feedback and ask us any queries you have

Schedule a call