Beat-Based Rhythm Quantization of MIDI Performances

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Wachter, Maximilian, Murgul, Sebastian, Heizmann, Michael
Format: Preprint
Published: 2025
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866915465447604224
author Wachter, Maximilian
Murgul, Sebastian
Heizmann, Michael
author_facet Wachter, Maximilian
Murgul, Sebastian
Heizmann, Michael
contents We propose a transformer-based rhythm quantization model that incorporates beat and downbeat information to quantize MIDI performances into metrically-aligned, human-readable scores. We propose a beat-based preprocessing method that transfers score and performance data into a unified token representation. We optimize our model architecture and data representation and train on piano and guitar performances. Our model exceeds state-of-the-art performance based on the MUSTER metric.
format Preprint
id arxiv_https___arxiv_org_abs_2508_19262
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle Beat-Based Rhythm Quantization of MIDI Performances
Wachter, Maximilian
Murgul, Sebastian
Heizmann, Michael
Sound
Computation and Language
Multimedia
Audio and Speech Processing
We propose a transformer-based rhythm quantization model that incorporates beat and downbeat information to quantize MIDI performances into metrically-aligned, human-readable scores. We propose a beat-based preprocessing method that transfers score and performance data into a unified token representation. We optimize our model architecture and data representation and train on piano and guitar performances. Our model exceeds state-of-the-art performance based on the MUSTER metric.
title Beat-Based Rhythm Quantization of MIDI Performances
topic Sound
Computation and Language
Multimedia
Audio and Speech Processing
url https://arxiv.org/abs/2508.19262