Quantifying Lexical Novelty in Song Lyrics

Novelty is an important psychological construct that affects both perceptual and behavioral processes. Here, we propose a lexical novelty score (LNS) for a song’s lyric, based on the statistical properties of a corpus of 275,905 lyrics (available at www.smcnus.org/lyrics/). A lyric-level LNS was derived as a function of the inverse document frequencies of its unique words. An artist-level LNS was then computed using the LNSs of lyrics uniquely associated with each artist. Statistical tests were performed to determine whether lyrics and artists on Billboard Magazine’s lists of “All-Time Top 100” songs and artists had significantly lower LNSs than “non-top” songs and artists. An affirmative and highly consistent answer was found in both cases. These results highlight the potential utility of the LNS as a feature for MIR.