Audio Features Affected by Music Expressiveness

TL;DR

This study investigates how musicians' emotional intentions influence audio features in recordings, using experiments with tuba players and analyzing the data through statistical and machine learning methods.

Contribution

It provides new insights into the relationship between musical expressiveness and audio features, combining traditional and machine learning analysis techniques.

Findings

Music expressiveness significantly affects audio features

Machine learning techniques can classify emotional content from audio features

Results inform future music information retrieval systems

Abstract

Within a Music Information Retrieval perspective, the goal of the study presented here is to investigate the impact on sound features of the musician's affective intention, namely when trying to intentionally convey emotional contents via expressiveness. A preliminary experiment has been performed involving $10$ tuba players. The recordings have been analysed by extracting a variety of features, which have been subsequently evaluated by combining both classic and machine learning statistical techniques. Results are reported and discussed.