cs.SDJul 31, 2026

Do Music Foundation Models Embed Pitch in Helical Structure?

Authors: Hayato YagiShinnosuke TakamichiRin SatoKeitaro TanakaShigeo Morishima

Organizations: Keio University, Japan · The University of Tokyo, Japan · Waseda University, Japan

Abstract

This study analyzes the intermediate representations of music foundation models (MFMs) and reports the geometric structures used to represent pitch information. By inputting isolated musical notes into trained MFMs and analyzing their principal components, we reveal that the representations form a helical structure reflecting the octave periodicity of pitch. Furthermore, we show that the clarity and geometry of this helical structure vary not only across models but also with the acoustic properties of the input signals. Our analysis provides a novel approach for clarifying the internal mechanisms of MFMs.

Explore similar work

CardsList
  1. PHALAR: Phasors for Learned Musical Audio Representations

    May 5, 2026Davide Marincione, Michele Mancusi, Giorgio Strano +4Sheet MusicRetrieval Task

  2. MuSViT: A Foundation Vision Model for Sheet Music Representation

    Jun 30, 2026Carlos Penarrubia, Antonio Rios-Vila, Eliseo Fuentes-Martinez +4Sheet MusicVision Foundation Models