Sellemain

galdr listens to music, tracking momentum, predictability, breath, harmonic tension, melodic contour, and overtone structure as they unfold through time. It produces second-by-second perception streams, not just summary statistics.