Models/config.yml

log_dir: "Models/Finetune_Extend"
save_freq: 1
log_interval: 5
device: "cuda"
epochs: 50 
batch_size: 3
max_len: 210 # maximum number of frames
pretrained_model: "Models/Finetune_Extend/current_model.pth"
load_only_params: false # set to true if do not want to load epoch numbers and optimizer parameters

data_params:
  train_data: "../../Data_Speech/viVoice/train.txt" 
  val_data: "../../Data_Speech/combine/combine_val.txt"
  root_path: "../../Data_Speech/"
  min_length: 50 # sample until texts with this size are obtained for OOD texts

preprocess_params:
  sr: 24000
  spect_params:
    n_fft: 2048
    win_length: 1200
    hop_length: 300

model_params:
  dim_in: 64 
  hidden_dim: 512
  max_conv_dim: 512
  n_layer: 3
  n_mels: 80

  n_token: 189 # number of phoneme tokens
  max_dur: 50 # maximum duration of a single phoneme
  style_dim: 128 # style vector size
  
  dropout: 0.2

  ASR_params:
    input_dim: 80
    hidden_dim: 256
    n_token: 189 # number of phoneme tokens
    n_layers: 6
    token_embedding_dim: 512

  JDC_params:
    num_class: 1
    seq_len: 192

  # config for decoder
  decoder: 
      type: 'hifigan' # either hifigan or istftnet
      resblock_kernel_sizes: [3,7,11]
      upsample_rates :  [10,5,3,2]
      upsample_initial_channel: 512
      resblock_dilation_sizes: [[1,3,5], [1,3,5], [1,3,5]]
      upsample_kernel_sizes: [20,10,6,4]
  
loss_params:
    lambda_mel: 5. # mel reconstruction loss
    lambda_gen: 1. # generator loss
    
    lambda_mono: 1. # monotonic alignment loss (TMA)
    lambda_s2s: 1. # sequence-to-sequence loss (TMA)

    lambda_F0: 1. # F0 reconstruction loss
    lambda_norm: 1. # norm reconstruction loss
    lambda_dur: 1. # duration loss
    lambda_ce: 20. # duration predictor probability output CE loss

optimizer_params:
  lr: 0.0001 # general learning rate
  ft_lr: 0.00001 # learning rate for acoustic modules