Dataset Information

A computational lens into how music characterizes genre in film.

ABSTRACT: Film music varies tremendously across genre in order to bring about different responses in an audience. For instance, composers may evoke passion in a romantic scene with lush string passages or inspire fear throughout horror films with inharmonious drones. This study investigates such phenomena through a quantitative evaluation of music that is associated with different film genres. We construct supervised neural network models with various pooling mechanisms to predict a film's genre from its soundtrack. We use these models to compare handcrafted music information retrieval (MIR) features against VGGish audio embedding features, finding similar performance with the top-performing architectures. We examine the best-performing MIR feature model through permutation feature importance (PFI), determining that mel-frequency cepstral coefficient (MFCC) and tonal features are most indicative of musical differences between genres. We investigate the interaction between musical and visual features with a cross-modal analysis, and do not find compelling evidence that music characteristic of a certain genre implies low-level visual features associated with that genre. Furthermore, we provide software code to replicate this study at https://github.com/usc-sail/mica-music-in-media. This work adds to our understanding of music's use in multi-modal contexts and offers the potential for future inquiry into human affective experiences.

SUBMITTER: Ma B

PROVIDER: S-EPMC8031455 | biostudies-literature |

REPOSITORIES: biostudies-literature

ACCESS DATA

Similar Datasets

Project description:Music making, in the form of free improvisations, is a common technique in music therapy, used to express one's feelings or ideas in the non-verbal language of music. In the broader sense, arts therapies, and music therapy in particular, are used to induce therapeutic and psychosocial effects, and to help mitigate symptoms in serious and chronic diseases. They are also used to empower the wellbeing and quality of life for both healthy individuals and patients. However, much research is still required to understand how music-based and arts-based approaches work, and to eventually enhance their effectivity. The clinical setting employing the arts constitutes a rich dynamic environment of occurrences that is difficult to capture, being driven by complex, simultaneous, and interwoven behavioral processes. Our computational paradigm is designed to allow substantial barriers in the arts-based fields to be overcome by enabling the rigorous and quantitative tracking, analyzing and documenting of the underlying dynamic processes. Here we expand the method for the music modality and apply it in a proof of principle experimentation to study expressive behavioral effects of diverse musical improvisation tasks on individuals and collectives. We have obtained statistically significant results that include empirical expressive patterns of feelings, as well as proficiency, gender and age behavioral differences, which point to variation factors of these categorized collectives in music making. Our results also suggest that males are more exploratory than females (e.g., they exhibit a larger range of octaves and intensity) and that the older people express musical characterized negativity more than younger ones (e.g., exhibiting larger note clusters and more chromatic transitions). We discuss implications of these findings to music therapy, such as behavioral diversity causality in treatment, as well as future scientific and clinical applications of the methodology.

Dataset Information

A computational lens into how music characterizes genre in film.

Similar Datasets

OmicsDI is part of the ELIXIR infrastructure

Tweets