Configs/config_ft.yml

log_dir: "Models/IMAS_FineTuned"
save_freq: 1
log_interval: 10
device: "cuda"
epochs: 50 # number of finetuning epoch (1 hour of data)
batch_size: 3
max_len: 2500 # maximum number of frames
pretrained_model: "/home/austin/disk2/llmvcs/tt/stylekan/Models/Style_Kanade/NO_SLM_3_epoch_2nd_00002.pth"
second_stage_load_pretrained: true # set to true if the pre-trained model is for 2nd stage
load_only_params: true # set to true if do not want to load epoch numbers and optimizer parameters

F0_path: "/home/austin/disk2/llmvcs/tt/stylekan/Utils/JDC/bst.t7"
ASR_config: "/home/austin/disk2/llmvcs/tt/stylekan/Utils/ASR/config.yml"
ASR_path: "/home/austin/disk2/llmvcs/tt/stylekan/Utils/ASR/bst_00080.pth"

PLBERT_dir: 'Utils/PLBERT/'

data_params:
  train_data: "/home/austin/disk2/llmvcs/tt/stylekan/Data/metadata_cleanest/FT_imas.csv"
  val_data: "/home/austin/disk2/llmvcs/tt/stylekan/Data/metadata_cleanest/FT_imas_valid.csv"
  root_path: ""
  OOD_data: "/home/austin/disk2/llmvcs/tt/stylekan/Data/OOD_LargeScale_.csv"
  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:
  multispeaker: true

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

  n_token: 178 # number of phoneme tokens
  max_dur: 50 # maximum duration of a single phoneme
  style_dim: 128 # style vector size
  
  dropout: 0.2

  decoder: 
    type: 'istftnet' # either hifigan or istftnet
    resblock_kernel_sizes: [3,7,11]
    upsample_rates :  [10, 6]
    upsample_initial_channel: 512
    resblock_dilation_sizes: [[1,3,5], [1,3,5], [1,3,5]]
    upsample_kernel_sizes: [20, 12]
    gen_istft_n_fft: 20
    gen_istft_hop_size: 5


      
  # speech language model config
  slm:
      model: 'Respair/Whisper_Large_v2_Encoder_Block' # The model itself is hardcoded, change it through -> losses.py
      sr: 16000 # sampling rate of SLM
      hidden: 1280 # hidden size of SLM
      nlayers: 33 # number of layers of SLM
      initial_channel: 64 # initial channels of SLM discriminator head
  
  # style diffusion model config
  diffusion:
    embedding_mask_proba: 0.1
    # transformer config
    transformer:
      num_layers: 3
      num_heads: 8
      head_features: 64
      multiplier: 2

    # diffusion distribution config
    dist:
      sigma_data: 0.2 # placeholder for estimate_sigma_data set to false
      estimate_sigma_data: true # estimate sigma_data from the current batch if set to true
      mean: -3.0
      std: 1.0
  
loss_params:
    lambda_mel: 10. # mel reconstruction loss
    lambda_gen: 1. # generator loss
    lambda_slm: 1. # slm feature matching loss
    
    lambda_mono: 1. # monotonic alignment loss (1st stage, TMA)
    lambda_s2s: 1. # sequence-to-sequence loss (1st stage, TMA)
    TMA_epoch: 9 # TMA starting epoch (1st stage)

    lambda_F0: 1. # F0 reconstruction loss (2nd stage)
    lambda_norm: 1. # norm reconstruction loss (2nd stage)
    lambda_dur: 1. # duration loss (2nd stage)
    lambda_ce: 20. # duration predictor probability output CE loss (2nd stage)
    lambda_sty: 1. # style reconstruction loss (2nd stage)
    lambda_diff: 1. # score matching loss (2nd stage)
    
    diff_epoch: 0 # style diffusion starting epoch (2nd stage)
    joint_epoch: 30 # joint training starting epoch (2nd stage)

optimizer_params:
  lr: 0.0001 # general learning rate
  bert_lr: 0.00001 # learning rate for PLBERT
  ft_lr: 0.00001 # learning rate for acoustic modules
  
slmadv_params:
  min_len: 400 # minimum length of samples
  max_len: 500 # maximum length of samples
  batch_percentage: 0.5 # to prevent out of memory, only use half of the original batch size
  iter: 20 # update the discriminator every this iterations of generator update
  thresh: 5 # gradient norm above which the gradient is scaled
  scale: 0.01 # gradient scaling factor for predictors from SLM discriminators
  sig: 1.5 # sigma for differentiable duration modeling