Circadian rhythm is a 24-hour cycle of human physiology that governs various aspects of human behavior and cognition, including attention, mood, and performance. This suggests that daily physiological rhythms may influence an individual's musical preferences and listening habits. The objective of this study is to investigate how musical preferences vary throughout the day and across seasons among the Indian population. To this end, we examined music consumption of the participants in terms of audio features by analyzing their one-year Spotify listening history to identify distinct diurnal patterns in music choices. Unsupervised clustering of musical features suggests that preference can be divided into four distinct time blocks: morning, afternoon, evening, and night. Besides, we extended the analysis by incorporating demographic information of the individuals under study. We also conducted a seasonal analysis to investigate potential differences in music consumption patterns between winter and summer months. It was found that individuals prefer music with positive valence, low arousal, and high danceability post-midnight. Additionally, loudness was recorded as the highest (maximum value) around 12 noon. Gender-based differences were also observed, with females perceiving music as slightly less energetic and less positive in terms of emotional valence compared to males, but showing a preference for higher acousticness. Age-based differences revealed that older individuals prefer music that is less energetic, less danceable, less loud, and more acoustic. These findings suggest that music preferences and perceptions vary based on gender and age, and that diurnal patterns play a role in shaping musical choices and listening habits. The findings can have significant implications for the music industry, particularly in creating tailored playlists and curating music to suit different times of the day. Furthermore, the analyses provide a framework for understanding the role of music in other cultures as well, allowing for cross-cultural comparisons, thereby showing how diurnal patterns and other physiological factors may influence culture-specific musical traditions.
Modern music notation software is extensive and so can be a comparative analysis. Since they employ a lot of different interactions to write scores, due to the mass of different symbols and their combinations, we developed a method to estimate the time spent by the user in interacting with the software interface in order to perform fundamental operations. For this we applied and extended the Keystroke-Level Model by analyzing interaction percentages in MusicXML files. Our findings contribute to modeling interaction and usability/ user experience research about interaction in music notation editors. These findings can be then transferred to analyze other editors and we expect to use the method in formative analyses to reduce user studies and thus development time in the long run.
In the performance of Western art music, musicians apply various strategies to manipulate the sound performed and communicate their musical interpretations through these subtle acoustic variations. It is a common practice for musicians to use typical conventions to express each compositional style (e.g., Baroque, Classical, or Romantic compositions). However, such stylistic expressive conventions have yet to be fully discussed in previous research. In this initial foray, we systematically compare the expressive strategies for different piano compositions. A series of piano performances are recorded in a controlled experimental setting (3 compositions × 8 pianists × 3 repeated trials = 72 recordings), and expressive acoustic elements are derived using Music Information Retrieval techniques. In our analysis, we reveal that expressive manners in music performance exhibit stable and systematic features corresponding to each music composition, and those stylistic trends serve as empirical observations for typical performance conventions in different music styles.
We propose a system to retrieve background music (BGM) for game scenes. BGM plays an important role in creating a particular atmosphere in game scenes, so studies have investigated the relationship between game scenes and BGM. However, none of the existing studies attempted to predict the audio features of BGM directly from a sequence of images expressing game scenes. In our system, the user inputs a sequence of images of a game scene, then our machine learning model, trained with gameplay videos, predicts the audio features from the input. Finally, the system retrieves the closest musical piece to the predicted audio features. Experimental results show both positive and negative tendencies: the predicted audio features for fight scenes are closer to the features of actually used BGM in fight scenes than those in other scenes (positive); the same musical piece was retrieved for different scenes (negative).
This paper describes a dataset of flute sounds with appropriate and inappropriate blowing styles. The flute is known as a difficult instrument to learn. We, therefore, have been developing a support system that automatically identifies the appropriateness of blowing in flute performances. To develop such a system, we need a dataset that consists of various sounds with various blowing styles, including both appropriate and inappropriate ways, but there are no such datasets. In this paper, we present the dataset that we have been developing. This dataset consists of sounds played by various players with various blowing styles, and also it has annotations of each sound's subjective evaluation.
