Estimating Musical Surprisal from Audio in Autoregressive Diffusion Model Noise Spaces
This work addresses the challenge of modeling musical expectancy and surprisal for audio analysis, presenting an incremental improvement over existing methods.
The paper tackles the problem of estimating musical surprisal from audio by using autoregressive diffusion models (ADMs) to calculate information content, showing that ADMs outperform a Generative Infinite-Vocabulary Transformer (GIVT) in negative log-likelihood and match or exceed it in tasks like capturing monophonic pitch surprisal and detecting segment boundaries.
Recently, the information content (IC) of predictions from a Generative Infinite-Vocabulary Transformer (GIVT) has been used to model musical expectancy and surprisal in audio. We investigate the effectiveness of such modelling using IC calculated with autoregressive diffusion models (ADMs). We empirically show that IC estimates of models based on two different diffusion ordinary differential equations (ODEs) describe diverse data better, in terms of negative log-likelihood, than a GIVT. We evaluate diffusion model IC's effectiveness in capturing surprisal aspects by examining two tasks: (1) capturing monophonic pitch surprisal, and (2) detecting segment boundaries in multi-track audio. In both tasks, the diffusion models match or exceed the performance of a GIVT. We hypothesize that the surprisal estimated at different diffusion process noise levels corresponds to the surprisal of music and audio features present at different audio granularities. Testing our hypothesis, we find that, for appropriate noise levels, the studied musical surprisal tasks' results improve. Code is provided on github.com/SonyCSLParis/audioic.