Upload 4 files
Browse files- config_ft.yml +21 -0
- epoch_2nd_00000.pth +3 -0
- epoch_2nd_00001.pth +3 -0
- epoch_2nd_00002.pth +3 -0
config_ft.yml
ADDED
|
@@ -0,0 +1,21 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{ASR_config: Utils/ASR/config.yml, ASR_path: Utils/ASR/epoch_00080.pth, F0_path: Utils/JDC/bst.t7,
|
| 2 |
+
PLBERT_dir: Utils/PLBERT/, batch_size: 2, data_params: {OOD_data: Data/OOD_texts.txt,
|
| 3 |
+
min_length: 50, root_path: Trelis_Data/wavs, train_data: Trelis_Data/train_list.txt,
|
| 4 |
+
val_data: Trelis_Data/val_list.txt}, device: cuda, epochs: 8, load_only_params: true,
|
| 5 |
+
log_dir: Models/LibriTTS-fft, log_interval: 10, loss_params: {diff_epoch: 2, joint_epoch: 4,
|
| 6 |
+
lambda_F0: 1.0, lambda_ce: 20.0, lambda_diff: 1.0, lambda_dur: 1.0, lambda_gen: 1.0,
|
| 7 |
+
lambda_mel: 5.0, lambda_mono: 1.0, lambda_norm: 1.0, lambda_s2s: 1.0, lambda_slm: 1.0,
|
| 8 |
+
lambda_sty: 1.0}, max_len: 600, model_params: {decoder: {resblock_dilation_sizes: [
|
| 9 |
+
[1, 3, 5], [1, 3, 5], [1, 3, 5]], resblock_kernel_sizes: [3, 7, 11], type: hifigan,
|
| 10 |
+
upsample_initial_channel: 512, upsample_kernel_sizes: [20, 10, 6, 4], upsample_rates: [
|
| 11 |
+
10, 5, 3, 2]}, diffusion: {dist: {estimate_sigma_data: true, mean: -3.0, sigma_data: 0.2473328872913062,
|
| 12 |
+
std: 1.0}, embedding_mask_proba: 0.1, transformer: {head_features: 64, multiplier: 2,
|
| 13 |
+
num_heads: 8, num_layers: 3}}, dim_in: 64, dropout: 0.2, hidden_dim: 512,
|
| 14 |
+
max_conv_dim: 512, max_dur: 50, multispeaker: true, n_layer: 3, n_mels: 80, n_token: 178,
|
| 15 |
+
slm: {hidden: 768, initial_channel: 64, model: microsoft/wavlm-base-plus, nlayers: 13,
|
| 16 |
+
sr: 16000}, style_dim: 128}, optimizer_params: {bert_lr: 1.0e-05, ft_lr: 0.0001,
|
| 17 |
+
lr: 0.0001}, preprocess_params: {spect_params: {hop_length: 300, n_fft: 2048,
|
| 18 |
+
win_length: 1200}, sr: 24000}, pretrained_model: Models/LibriTTS/epochs_2nd_00020.pth,
|
| 19 |
+
save_freq: 1, second_stage_load_pretrained: true, slmadv_params: {batch_percentage: 1,
|
| 20 |
+
iter: 10, max_len: 500, min_len: 400, scale: 0.01, sig: 1.5, thresh: 5}, wandb_project: StyleTTS2,
|
| 21 |
+
wandb_run: diff2-adv4-fft-600fr-2b-8ep-trelis_voice_512_V5-ronan}
|
epoch_2nd_00000.pth
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:86a4ac7f2fbeec155e958ef4e6f68e6d8dac9ad806cd2ecb60608d5bc433e75b
|
| 3 |
+
size 2040174838
|
epoch_2nd_00001.pth
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:a8d80b870cd202b50884b908becd68eace9c4869bd170b5cfac324a81722f66f
|
| 3 |
+
size 2040174838
|
epoch_2nd_00002.pth
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:4e3e8329dd00a04c3bae1afb4d97cacad97c5a4a91c0519dd24a7cfb65b9c84b
|
| 3 |
+
size 2242832422
|