Cross-dataset music emotion recognition: an end-to-end approach

Abstract

The topic of Music Emotion Recognition (MER) evolved as music is a fascinating expression of emotions, yet it faces challenges given its subjectivity. Because each language has its particularities in terms of sound and intonation, and implicitly associations made upon them, we hypothesize perceived emotions might vary in different cultures. To address this issue, we test a novel approach towards emotion detection and propose a language sensitive end-to-end model that learns to tag emotions from music with lyrics in English, Mandarin and Turkish.