vocab_size: 6001
embedding_dim: 1024

in_channels: 1024
out_channels: 1
resblock_type: "1"
resblock_dilation_sizes: [[1, 3, 5], [1, 3, 5], [1, 3, 5]]
resblock_kernel_sizes: [3, 7, 11]
upsample_kernel_sizes: [11, 8, 8, 4, 4]
upsample_initial_channel: 512
upsample_factors: [5, 4, 4, 2, 2]
inference_padding: 5
cond_channels: 0
conv_post_bias: True

generator: !new:speechbrain.lobes.models.HifiGAN.UnitHifiganGenerator
  in_channels: !ref <in_channels>
  out_channels: !ref <out_channels>
  resblock_type: !ref <resblock_type>
  resblock_dilation_sizes: !ref <resblock_dilation_sizes>
  resblock_kernel_sizes: !ref <resblock_kernel_sizes>
  upsample_kernel_sizes: !ref <upsample_kernel_sizes>
  upsample_initial_channel: !ref <upsample_initial_channel>
  upsample_factors: !ref <upsample_factors>
  inference_padding: !ref <inference_padding>
  cond_channels: !ref <cond_channels>
  conv_post_bias: !ref <conv_post_bias>
  vocab_size: !ref <vocab_size>
  embedding_dim: !ref <embedding_dim>
  duration_predictor: False
  multi_speaker: False
  skip_token_embedding: True

modules:
  generator: !ref <generator>

pretrainer: !new:speechbrain.utils.parameter_transfer.Pretrainer
    loadables:
        generator: !ref <generator>