Abstract
Melody prediction is an important aspect of music listening. The success of prediction, i.e., whether the next note played in a song is the same as the one predicted by the listener, depends on various factors. In the paper, we present two studies, where we assess how music familiarity and music expertise influence melody prediction in human listeners, and, expressed in appropriate data/algorithmic ways, computational models. To gather data on human listeners, we designed a melody prediction user study, where familiarity was controlled by two different music collections, while expertise was assessed by adapting the Music Sophistication Index instrument to Slovenian language. In the second study, we evaluated the melody prediction accuracy of computational melody prediction models. We evaluated two models, the SymCHM and the Implication-Realization model, which differ substantially in how they approach melody prediction. Our results show that both music familiarity and expertise affect the prediction accuracy of human listeners, as well as of computational models.
Highlights
One of the main aspects of listening to music is the tendency of the brain to constantly predict the upcoming melodic events
We evaluated the performance of two algorithms, (i) the Implication-Realization (I-R) model developed by Narmour (1990), which is agnostic of musical culture, and (ii) the Compositional Hierarchical Model for symbolic music representations (SymCHM) developed by Pesek et al (2017b), which is trained on a dataset of songs and biased toward familiar songs
We present two studies on how melody prediction is affected by music familiarity and expertise in (1) human listeners and (2) computational approaches
Summary
One of the main aspects of listening to music is the tendency of the brain to constantly predict the upcoming melodic events. How human listeners perform the ongoing prediction of music is influenced by (i) their general music expertise and by (ii) their familiarity with the type of music they are listening to. These two concepts are two facets of the knowledge that listeners possess. Research on understanding human melody prediction has crossed over to the development of computational models that perform melody prediction The knowledge that such models use, typically stems from a dataset that the researchers develop or train their models on. The question is how do models, trained on a dataset, and humans, trained through years of listening to music, compare in terms of melody prediction.
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
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 callDisclaimer: All third-party content on this website/platform is and will remain the property of their respective owners and is provided on "as is" basis without any warranties, express or implied. Use of third-party content does not indicate any affiliation, sponsorship with or endorsement by them. Any references to third-party content is to identify the corresponding services and shall be considered fair use under The CopyrightLaw.
