Embeddings as representation for symbolic music

TL;DR

This paper explores the use of embeddings to encode musical notes, aiming to capture musical meaning and improve models for tasks like melody and harmony generation, inspired by NLP techniques.

Contribution

It introduces a method to represent musical notes with embeddings and analyzes their effectiveness in capturing musical patterns across datasets.

Findings

Embeddings can encode meaningful musical patterns.

t-SNE visualizations reveal structure in musical note embeddings.

Potential for improved music generation models.

Abstract

A representation technique that allows encoding music in a way that contains musical meaning would improve the results of any model trained for computer music tasks like generation of melodies and harmonies of better quality. The field of natural language processing has done a lot of work in finding a way to capture the semantic meaning of words and sentences, and word embeddings have successfully shown the capabilities for such a task. In this paper, we experiment with embeddings to represent musical notes from 3 different variations of a dataset and analyze if the model can capture useful musical patterns. To do this, the resulting embeddings are visualized in projections using the t-SNE technique.