Towards Robust Unsupervised Disentanglement of Sequential Data -- A Case Study Using Music Audio. (arXiv:2205.05871v1 [cs.SD])

Disentangled sequential autoencoders (DSAEs) represent a class of
probabilistic graphical models that describes an observed sequence with dynamic
latent variables and a static latent variable. The former encode information at
a frame rate identical to the observation, while the latter globally governs
the entire sequence. This introduces an inductive bias and facilitates
unsupervised disentanglement of the underlying local and global factors. In
this paper, we show that the vanilla DSAE suffers from being sensitive to the
choice of model architecture …

arxiv audio case case study data music study unsupervised