<p>Noncollinear spin structures are promising platforms for non-volatile information storage. Controlling these textures through magnetic excitations enables ultrafast and low-dissipation signal processing, offering a route toward next-generation spintronic technology. To simulate the interplay between excited state and spin structure, we present a first-principles approach for computing magnon in magnetic systems with noncollinear spin textures, based on density functional theory and many-body perturbation theory. The computational challenges posed by nanoscale spin configurations—often tens to thousands of times larger than a primitive cell—are overcome through a Wannier-basis representation and a data-driven ansatz potential method, enabling parameter-free simulations of systems containing O(10²) spins, in stark contrast with O(10⁰) spins of typical collinear cases. We apply our method to the spin-spiral state of type-II multiferroic LiCu₂O₂, successfully capturing its steady-state spin-rotation pitch and resolving the magnon dispersion in agreement with the experimental measurements. We further analyze the interplay between the spiral spin structure and the spin-exchange splitting, elucidating the crucial role of magnetic dipoles on ligand ions in mediating effective ferromagnetic interaction among the primary spins on Cu²⁺ ions. Overall, this work establishes a general and computationally efficient framework for simulating collective spin dynamics in large-scale noncollinear magnetic systems from first principles.</p>