Modeling Temporal Dependencies in High-Dimensional Sequences: Application to Polyphonic Music Generation and Transcription

TL;DR

This paper presents a probabilistic neural network model that captures temporal dependencies in high-dimensional polyphonic music sequences, improving generation and transcription accuracy.

Contribution

It introduces a novel RNN-based distribution estimator for modeling complex polyphonic music sequences, outperforming traditional models.

Findings

Outperforms traditional models on realistic datasets

Enhances polyphonic transcription accuracy

Effectively captures temporal dependencies in high-dimensional data

Abstract

We investigate the problem of modeling symbolic sequences of polyphonic music in a completely general piano-roll representation. We introduce a probabilistic model based on distribution estimators conditioned on a recurrent neural network that is able to discover temporal dependencies in high-dimensional sequences. Our approach outperforms many traditional models of polyphonic music on a variety of realistic datasets. We show how our musical language model can serve as a symbolic prior to improve the accuracy of polyphonic transcription.