Learning to Traverse Latent Spaces for Musical Score Inpainting

TL;DR

This paper introduces a deep learning model that uses latent space traversal to perform musical score inpainting, effectively filling in missing music segments by considering both past and future contexts for interactive music creation.

Contribution

The paper presents a novel approach combining Variational Auto-Encoder and Recurrent Neural Network for musically meaningful score inpainting, advancing interactive music creation tools.

Findings

The model generates coherent musical inpaintings connecting musical excerpts.

It outperforms baseline methods in objective and subjective evaluations.

Demonstrates the effectiveness of latent space traversal in deep generative models for music.

Abstract

Music Inpainting is the task of filling in missing or lost information in a piece of music. We investigate this task from an interactive music creation perspective. To this end, a novel deep learning-based approach for musical score inpainting is proposed. The designed model takes both past and future musical context into account and is capable of suggesting ways to connect them in a musically meaningful manner. To achieve this, we leverage the representational power of the latent space of a Variational Auto-Encoder and train a Recurrent Neural Network which learns to traverse this latent space conditioned on the past and future musical contexts. Consequently, the designed model is capable of generating several measures of music to connect two musical excerpts. The capabilities and performance of the model are showcased by comparison with competitive baselines using several objective and subjective evaluation methods. The results show that the model generates meaningful inpaintings and can be used in interactive music creation applications. Overall, the method demonstrates the merit of learning complex trajectories in the latent spaces of deep generative models.