This note presents a system that learns expressive and idiosyncratic gesture variations for gesture-based interaction. The system is used as an interaction technique in a music conducting scenario where gesture variations drive music articulation. A simple model based on Gaussian Mixture Modeling is used to allow the user to configure the system by providing variation examples. The system performance and the influence of user musical expertise is evaluated in a user study, which shows that the model is able to learn idiosyncratic variations that allow users to control articulation, with better performance for users with musical expertise.

Investigations on music and emotion have identified broad musical elements that influence emotions recognized by listeners, such as timbre, rhythm, melody, and harmony. Not many studies have studied the correlation between quantifiable musical descriptors and their associated emotions; furthermore, only few studies have focused on how listeners’ demographic and musical backgrounds influence the emotion they recognize. In this preliminary study, participants rated how strongly they recognized the six GEMS emotions (transcendence, peacefulness, power, joyful activation, tension, and sadness) while listening to excerpts from Beethoven’s Eroica. Musical descriptors (loudness, brightness, noisiness, tempo/rhythm, harmony, and timbre) were also extracted from each excerpt. Results indicate significant correlations between emotional ratings and musical descriptors, notably positive correlations between key clarity and peacefulness/joyful activation ratings, and negative correlations between key clarity and tension/sadness ratings. Key clarity refers to the key strength associated to the best key candidate; as such, these results suggest that listeners recognize positive emotions in music with a straightforward key, whereas listeners recognize negative emotions in music with a less clear sense of key. The second part of the study computed correlations between demographics and emotional ratings, to determine whether people of similar demographic and musical backgrounds recognized similar emotions. The results indicate that na{\"ıve listeners (i.e. younger subjects, and subjects with less frequent exposure to classical music) experienced more similar emotions from the same musical excerpts than did other subjects. Our findings contribute to developing a quantitative understanding of how musical descriptors, and listeners’ backgrounds, correlate with emotions recognized by listeners.

In this paper we describe our preliminary work on understanding the impact of personality on the emotion induction in dierent social circumstances during the consumption of movies, for the purposeof the context-aware recommender system for movies. The purpose ofthis study is to answer two research questions: is there a dierence in emotion induction when users are alone as opposed to when they are with company during watching the movie, and do dierent personality proles in uence the emotion induction when users are alone as opposed to when they are with company during watching the movie? We have used the (LDOS-CoMoDa) dataset which contains ratings and associated contextual information for the consumed movies, as well as Big Five personality proles of the users. The results showed that there is an in uence of social context on emotion induction, and that personality factors have to be taken into consideration since for the dierent groups of users, based on the personality factors, the emotion induction was in uenced dierently