0.1.8

address https://github.com/lucidrains/naturalspeech2-pytorch/issues/29

0.1.7

condition latent features with aligned conditions prior to wavenet stack

0.1.6

0.1.6

0.1.5

fix amp

0.1.4

account for variable sequence lengths when generating alignment mask …

…from durations

0.1.2

just make a guess about pyworld and make sure it runs

0.1.1

make sure forward sum loss actually runs, cleanup

0.1.0

0.1.0

0.0.50

0.0.50

0.0.49

at inference time, the alignment mask is derived from the duration. i…

…mprovise a get_mask_from_lengths function, consult with someone in the field later