final model release

final model for 32k

48k 5 layers model

maxgan_pretrain_48K_5L.pth is based on maxgan_pretrain_16K_5L.pth by transfer learning, so the struct maybe not be the most suitable for 48K.

16K 5 layers model

upsample_rates: [5,4,2,2,2]
upsample_kernel_sizes: [15,12,4,4,4]
upsample_initial_channel: 512

maxgan v1 pretrain model, new best

使用噪声增广数据进行训练的maxgan_pretrain.pth，包含了generator&discriminator

maxgan v1 pretrain model

正式发布版本：maxgan_pretrain.pth 中包含了generator&discriminator

preview 56 singers model of opensinger data

preview model is trained of 415 epochs, the 1000 epochs model will be released some days latter.

baker model

train on pure speech, 340 epochs, 10k steps

opencpop model

train on one T4 about 5 days, 1960 epoch

preview_model

基于opencpop音色的预览模型