Search papers, labs, and topics across Lattice.
The paper introduces the Dissonance Spectrum (DS), a novel nonnegative time-frequency representation that models perceptual frequency interactions using a tolerance-based rational pitch-relation kernel. By applying this approach to conventional music representations, the authors achieve strong ordinal agreement with music-theory constructs, outperforming baseline models in music question answering and emotion recognition tasks. The findings suggest that DS provides a more interpretable and effective framework for understanding music, although challenges in listener-specific perception and task coverage remain.
The Dissonance Spectrum reveals intricate frequency interactions that conventional models overlook, leading to superior performance in music understanding tasks.
Conventional music representations describe acoustic energy over time and frequency but do not explicitly expose relations among simultaneous frequency components. We introduce the \emph{Dissonance Spectrum} (DS), a nonnegative time--frequency representation that applies a tolerance-based rational pitch-relation kernel with logarithmic harmonic distance to a constant-Q spectrum and attributes aggregate pairwise interactions back to individual frequency bins. Controlled music-theory tests show strong ordinal agreement for intervals, harmonic-function connections, and church modes, and weaker but significant agreement across diverse chord voicings. DS is then encoded by a lightweight parallel branch whose zero-initialized residual projection preserves the baseline function at initialization. Across six paired training seeds in open-ended music question answering and categorical and dimensional music emotion recognition, DS obtains the highest mean on every reported endpoint relative to the unchanged baseline, a parameter-matched Gaussian-input branch, and an architecture-matched magnitude-CQT branch. These results support DS as an interpretable, complementary representation, while listener-specific perception and broader task coverage remain open problems.
