params.py

from model.utils import fix_len_compatibility


# data parameters
train_filelist_path = 'fs2_txt/train.txt' #same split as in the FastSpeech2 paper
valid_filelist_path = 'fs2_txt/valid.txt'
test_filelist_path = 'fs2_txt/test.txt'
cmudict_path = 'resources/cmu_dictionary'
add_blank = True
n_feats = 80
n_spks = 1   
spk_emb_dim = 64
n_feats = 80
n_fft = 1024
sample_rate = 22050
hop_length = 256
win_length = 1024
f_min = 0
f_max = 8000

# encoder parameters
n_enc_channels = 192
filter_channels = 768
filter_channels_dp = 256
n_enc_layers = 6
enc_kernel = 3
enc_dropout = 0.1
n_heads = 2
window_size = 4

teacher = False # true for teacher model, false for consistency distillation

 
# training parameters
log_dir = 'logs/20230826_student' 
test_size = 2
n_epochs = 10000
batch_size =  16  
learning_rate = 1e-4
seed = 1234
save_every = 10 
out_size =  fix_len_compatibility(2*22050//256)