The Use of Data-Based Emotion Recognition Systems for the Development of Emotional and Musical Creativity in Vocal Students

The primary aim of this study was to evaluate whether the use of the EmoMusic dataset could enhance first-year students’ emotional and musical creativity. A randomized controlled trial was conducted with 128 students. Participants were randomly assigned either to a control group that continued traditional vocal training or to an experimental group that incorporated the EmoMusic dataset into the curriculum. The experimental group utilized the EmoMusic system for 30 minutes at the end of each session, twice a week over a three-month period. Creativity in music was assessed using the Thinking Creatively with Sounds and Words (TCSW) tool, while emotional creativity was evaluated using the Emotional Creativity Inventory (ECI). Data were analyzed using non-parametric statistical tests, specifically the Wilcoxon Signed-Rank and Mann-Whitney U test. Secondary statistical analysis revealed statistically significant differences between the control and experimental groups following the intervention, with p -values <0.001, indicating the effectiveness of the proposed intervention. The Wilcoxon Signed-Rank Test also significantly improved the experimental group across all measured parameters ( p < .001). These results suggest that the incorporation of neural network-based emotion recognition systems into vocal training programs can be effective in enhancing creativity and emotional expressiveness.