A Neural Score Follower for Computer Accompaniment of Polyphonic Musical Instruments

2025

Read A Neural Score Follower for Computer Accompaniment of Polyphonic Musical Instruments (2025) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/arxiv-2503-06348.txt.

Description

Ashwin Pillay builds a neural score follower for live computer accompaniment of musicians. A score follower tracks where a player sits in the written music in real time. Past tools used dynamic programming, hidden Markov models, or time warping for that task. Deep learning had not been tried directly on the problem before this work.

His system, HeurMiT, learns squeezed codes of the music that hold the player's place. A MIDI augmentation kit toughens those codes against player slips. Heuristic rules join the network to link with transcription and accompaniment tools. Tests showed blazing speed but deep limits in real use. Pillay frames the work as a first probe and maps paths toward a sound neural follower.

Catalog record

YEAR2025
COLLECTIONPAPERS
FORMATTXT
FILE SIZE83 KB
VERSION1.0
LICENSEhttp://creativecommons.org/licenses/by/4.0/
RIGHTSCC BY
SOURCEarXiv

Source: https://arxiv.org/abs/2503-06348

Browse the full library · About the Tower of Babel