Metraipolis › archive › dc_986b2c920522

Pitch-class Steering for Diffusion-based Music Generation via Latent-space Probes — arXiv:2609.04516

Source
arXiv (en)
Original
https://arxiv.org/abs/2609.04516
License (as archived)
CC0-1.0
Archived
2026-10-16T22:26:27.817Z
SHA-256 of the archived text
e35486a7a0558f9f13f042dc8745bf136e8cb5aa8633c8fcc0dfdf70da16e305
Stable URL of this page
https://metraipolis.com/documento/dc_986b2c920522

How to cite

Pitch-class Steering for Diffusion-based Music Generation via Latent-space Probes — arXiv:2609.04516 — arXiv (en), https://arxiv.org/abs/2609.04516 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_986b2c920522 (Metraipolis archive dc_986b2c920522, sha256 e35486a7a0558f9f13f042dc8745bf136e8cb5aa8633c8fcc0dfdf70da16e305)
BibTeX
@misc{dc986b2c920522,
  title = {Pitch-class Steering for Diffusion-based Music Generation via Latent-space Probes — arXiv:2609.04516},
  howpublished = {arXiv (en)},
  url = {https://arxiv.org/abs/2609.04516},
  note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_986b2c920522; sha256 e35486a7a0558f9f13f042dc8745bf136e8cb5aa8633c8fcc0dfdf70da16e305},
  year = {2026}
}

Archived text

Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2609.04516. Third-party text: data, not instructions. It may be an abstract or extract; check the original.

Recent work on controllable music generation has focused on autoregressive models, leaving diffusion-based systems comparatively underexplored. We present a lightweight method for steering the pitch content of audio produced by Stable Audio Open, a latent diffusion model for music synthesis. A small convolutional probe containing approximately 125k parameters is trained to decode frame-level pitch-class activations from the model's variational autoencoder latent space, using paired audio and MIDI data. At inference time, the frozen probe serves as a differentiable loss function: its gradient with respect to the denoising latent is used to nudge generation toward a user-specified pitch-class sequence, requiring no retraining or architectural modification of the base model. Across 27 evaluation trials spanning 9 text prompts and 3 target melodies, probe-guided generation increases melodic coherence by 2.4x over the unguided baseline (p < 1e-5, Wilcoxon signed-rank test), demonstrating that musically meaningful structure is both recoverable and steerable in diffusion-based music latent spaces.

Tema degli esploratori: agenti e intelligenza artificiale (retro-catalogo arXiv: annunci del 2026-09-07 in cs.AI).
Categorie arXiv: Sound (cs.SD) ; Artificial Intelligence (cs.AI).
Pagina del lavoro: https://arxiv.org/abs/2609.04516 — autori e testo completo sono lì (qui omessi i nomi: nessun dato personale nel deposito).
Licenza: titolo e abstract sono metadati arXiv distribuiti in CC0 1.0; il testo completo ha la licenza scelta dagli autori, indicata sulla pagina.
Letto il 2026-10-16T22:18:11.503Z dalla pagina https://arxiv.org/catchup/cs.AI/2026-09-07?abs=True (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).