Deep Generative Models of Music Expectation

TL;DR

This paper introduces a deep diffusion probabilistic model to estimate musical surprisal, demonstrating its ability to predict human liking ratings more accurately than previous methods, advancing understanding of music expectation.

Contribution

The work applies deep diffusion models to learn complex musical features and predict surprisal, outperforming traditional models like IDyOM in correlating with human preferences.

Findings

Diffusion models produce surprisal values with a negative quadratic relation to liking.

The model's performance is competitive with state-of-the-art methods.

Deep neural networks can learn complex features directly from musical data.

Abstract

A prominent theory of affective response to music revolves around the concepts of surprisal and expectation. In prior work, this idea has been operationalized in the form of probabilistic models of music which allow for precise computation of song (or note-by-note) probabilities, conditioned on a 'training set' of prior musical or cultural experiences. To date, however, these models have been limited to compute exact probabilities through hand-crafted features or restricted to linear models which are likely not sufficient to represent the complex conditional distributions present in music. In this work, we propose to use modern deep probabilistic generative models in the form of a Diffusion Model to compute an approximate likelihood of a musical input sequence. Unlike prior work, such a generative model parameterized by deep neural networks is able to learn complex non-linear features directly from a training set itself. In doing so, we expect to find that such models are able to more accurately represent the 'surprisal' of music for human listeners. From the literature, it is known that there is an inverted U-shaped relationship between surprisal and the amount human subjects 'like' a given song. In this work we show that pre-trained diffusion models indeed yield musical surprisal values which exhibit a negative quadratic relationship with measured subject 'liking' ratings, and that the quality of this relationship is competitive with state of the art methods such as IDyOM. We therefore present this model a preliminary step in developing modern deep generative models of music expectation and subjective likability.