Nonnegative Tucker Decomposition with Beta-divergence for Music Structure Analysis of Audio Signals. (arXiv:2110.14434v4 [cs.SD] UPDATED)

Nonnegative Tucker decomposition (NTD), a tensor decomposition model, has
received increased interest in the recent years because of its ability to
blindly extract meaningful patterns, in particular in Music Information
Retrieval. Nevertheless, existing algorithms to compute NTD are mostly designed
for the Euclidean loss. This work proposes a multiplicative updates algorithm
to compute NTD with the beta-divergence loss, often considered a better loss
for audio processing. We notably show how to implement efficiently the
multiplicative rules using tensor algebra. Finally, …

analysis arxiv audio beta divergence music