Towards Explaining Expressive Qualities in Piano Recordings: Transfer of Explanatory Features via Acoustic Domain Adaptation

Abstract

Emotion and expressivity in music have been topics of considerable interest in the field of music information retrieval. In recent years, mid-level perceptual features have been suggested as means to explain computational predictions of musical emotion. We find that the diversity of musical styles and genres in the available dataset for learning these features is not sufficient for models to generalise well to specialised acoustic domains such as solo piano music. In this work, we show that by utilising unsupervised domain adaptation together with receptive-field regularised deep neural networks, it is possible to significantly improve generalisation to this domain. Additionally, we demonstrate that our domain-adapted models can better predict and explain expressive qualities in classical piano performances, as perceived and described by human listeners.

Tracing Back Music Emotion Predictions to Sound Sources and Intuitive Perceptual Qualities

“…An immediate next step that we are currently pursuing is to extend audioLIME to provide explanations in the form of temporal segments using semantic music segmentation, along with the sound sources. We are also looking at explaining emotion conveyed in classical piano performances, which pose particular challenges -including the non-availability of training data, where transfer learning of explanatory features becomes necessary [32].…”