SCORE-SET: A dataset of GuitarPro files for Music Phrase Generation and Sequence Learning

TL;DR

SCORE-SET is a specialized GuitarPro dataset designed for music generation and sequence learning, incorporating expressive guitar performance features to enhance realism in machine learning models.

Contribution

The paper introduces a new, curated dataset of GuitarPro files with detailed expressive annotations, tailored for guitar music generation and sequence modeling tasks.

Findings

Dataset includes diverse guitar expression features.

Enables improved performance in guitar music generation models.

Facilitates research in performance-aware music synthesis.

Abstract

A curated dataset of Guitar Pro tablature files (.gp5 format), tailored for tasks involving guitar music generation, sequence modeling, and performance-aware learning is provided. The dataset is derived from MIDI notes in MAESTRO and GiantMIDI which have been adapted into rhythm guitar tracks. These tracks are further processed to include a variety of expression settings typical of guitar performance, such as bends, slides, vibrato, and palm muting, to better reflect the nuances of real-world guitar playing.