Smart personalized music recommendation and playlist generation 

Humans are both listeners and music-makers and throughout history, in every part of the world and every culture, music has constituted a universal feature, partly owing to its power to evoke strong emotions and influence moods. As humans listen to music daily, the sentiment of music being listened to can not only reflect the mood of the listener but it can also alter it. This is of even higher significance nowadays, since the unprecedented challenges caused by the COVID-19 pandemic have led to more stress and anxiety than ever before. Given the severe limitations in interpersonal interactions and mobility, it is now of higher relevance and importance to provide the most suitable and personalized tools that will be able to assist us in any way possible.

i-MRePlay proposes the novel combination of state-of-the-art and beyond signal processing, machine and deep learning techniques for the analysis of music based on emotion, genre, lyrics and other music-related parameters, such as instrumentation (music source separation techniques will be utilized to be able to recommend music, based purely on specific instrumental properties), for the recommendation of smart personalized music playlists; a research field of increased interest during the last years. Since people’s personality variations play a crucial role in both emotion elicitation and music preferences, personalization techniques will be examined. Thus, an analysis on how COVID-19, as a case study of a consequential real world event, has affected people’s musical preferences will be conducted, in order to quantify factors that express the potential sentiment shift and examine how music could be possibly used to alter mood.

This research is supported by the Hellenic Foundation for Research and Innovation (H.F.R.I.) under the “3rd Call for H.F.R.I. Research Projects to support Post-Doctoral Researchers” (Project Number: 7773).