Releases: archinetai/audio-diffusion-pytorch
Releases · archinetai/audio-diffusion-pytorch
v0.0.71
feat: add tanh bottleneck, option to use multiple bottlenecks
v0.0.70
feat: add option to use rel pos
v0.0.69
fix: v-sampler output
v0.0.68
fix: add linear schedule in init
v0.0.67
feat: add true v-diffusion, separate vk-diffusion
v0.0.66
fix: forward context_features, big bug that prevented conditioning fr…
v0.0.65
feat: add mean and logvar info to variational bottleneck
v0.0.64
fix: t5 embedder device
v0.0.63
fix: change to auto tokenizer
v0.0.62
feat: add pretrained/frozen T5 embedder