Studying induced musical emotion via a corpus of annotations collected through crowd‐sourcing

Abstract

One of the major reasons why music is so enjoyable is its emotional impact. For many people, music is an important everyday aid of emotional regulation. As such, music is used by musical therapists to and in entertainment industry. Recently, mechanisms of emotional induction through music received a lot of attention from different scientific fields, including musicology, neuroscience, psychology, and computer science [1]. In this study we present a computer-science approach to this problem, comprising crowd-sourcing for data collection and a size and structure of the dataset that allows for building a computational model.