Pitch-class Steering for Diffusion-based Music Generation via Latent-space Probes

The paper introduces a lightweight technique to steer the pitch content of audio generated by the Stable Audio Open diffusion model. A small convolutional probe (~125k parameters) is trained to decode frame‑level pitch‑class activations from the model’s latent space using paired audio and MIDI data. During inference, the frozen probe acts as a differentiable loss, guiding generation toward a user‑specified pitch‑class sequence without retraining the base model, and improves melodic coherence by 2.4× over the unguided baseline.

Machine-generated by The Flow from the publisher's headline and feed description — not written or checked by a human. The full article lives at arXiv AI.

• diffusion