1.6.4

address https://github.com/lucidrains/audiolm-pytorch/issues/242

1.6.3

allow for local attention to attend to nothing

1.6.2

increase default codebook size

1.6.1

allow for grouped residual LFQ

1.6.0

integrate residual lookup free quantization

1.5.7

address padding in acoustic tokens when sampling from fine transforme…

1.5.6

address https://github.com/lucidrains/audiolm-pytorch/issues/235

1.5.5

more barriers

1.5.4

fix validation in trainers

1.5.3

allow for setting init process timeout as well as an extra barrier