rippertnt commited on
Commit
1363ef0
1 Parent(s): d7a42b0

Upload 15 files

Browse files
Files changed (12) hide show
  1. fa_base.json +62 -0
  2. fa_base.onnx +3 -0
  3. fa_base.pth +3 -0
  4. km_base.json +62 -0
  5. km_base.onnx +3 -0
  6. km_base.pth +3 -0
  7. mn_base.json +62 -0
  8. mn_base.onnx +3 -0
  9. mn_base.pth +3 -0
  10. tr_base.json +62 -0
  11. tr_base.onnx +3 -0
  12. tr_base.pth +3 -0
fa_base.json ADDED
@@ -0,0 +1,62 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "train": {
3
+ "log_interval": 1000,
4
+ "eval_interval": 5000,
5
+ "seed": 1234,
6
+ "epochs": 128,
7
+ "learning_rate": 1e-4,
8
+ "betas": [0.8, 0.99],
9
+ "eps": 1e-9,
10
+ "batch_size": 8,
11
+ "fp16_run": false,
12
+ "lr_decay": 0.999875,
13
+ "segment_size": 8192,
14
+ "init_lr_ratio": 1,
15
+ "warmup_epochs": 0,
16
+ "c_mel": 45,
17
+ "c_kl": 1.0
18
+ },
19
+ "data": {
20
+ "use_mel_posterior_encoder": true,
21
+ "training_files":"filelists/fa_train.txt.cleaned",
22
+ "validation_files":"filelists/fa_valid.txt.cleaned",
23
+ "text_cleaners":["canvers_fa_cleaners"],
24
+ "max_wav_value": 32768.0,
25
+ "sampling_rate": 22050,
26
+ "filter_length": 1024,
27
+ "hop_length": 256,
28
+ "win_length": 1024,
29
+ "n_mel_channels": 80,
30
+ "mel_fmin": 0.0,
31
+ "mel_fmax": null,
32
+ "add_blank": false,
33
+ "n_speakers": 2,
34
+ "cleaned_text": true
35
+ },
36
+ "model": {
37
+ "use_mel_posterior_encoder": true,
38
+ "use_transformer_flows": true,
39
+ "transformer_flow_type": "pre_conv",
40
+ "use_spk_conditioned_encoder": true,
41
+ "use_noise_scaled_mas": true,
42
+ "use_duration_discriminator": true,
43
+ "inter_channels": 192,
44
+ "hidden_channels": 192,
45
+ "filter_channels": 768,
46
+ "n_heads": 2,
47
+ "n_layers": 6,
48
+ "kernel_size": 3,
49
+ "p_dropout": 0.1,
50
+ "resblock": "1",
51
+ "resblock_kernel_sizes": [3,7,11],
52
+ "resblock_dilation_sizes": [[1,3,5], [1,3,5], [1,3,5]],
53
+ "upsample_rates": [8,8,2,2],
54
+ "upsample_initial_channel": 512,
55
+ "upsample_kernel_sizes": [16,16,4,4],
56
+ "n_layers_q": 3,
57
+ "use_spectral_norm": false,
58
+ "use_sdp": false,
59
+ "gin_channels": 256
60
+ }
61
+ }
62
+
fa_base.onnx ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:c75f22f87ae602a82d128a248f6d326cd873ff3fc88f8efa5236dd3d8c5ccff9
3
+ size 123427901
fa_base.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:21af5b11b3a1a68e0a72999fe9b062bb39fd969948687804baf39986e1442742
3
+ size 485792475
km_base.json ADDED
@@ -0,0 +1,62 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "train": {
3
+ "log_interval": 1000,
4
+ "eval_interval": 5000,
5
+ "seed": 1234,
6
+ "epochs": 128,
7
+ "learning_rate": 2e-5,
8
+ "betas": [0.8, 0.99],
9
+ "eps": 1e-9,
10
+ "batch_size": 12,
11
+ "fp16_run": false,
12
+ "lr_decay": 0.999875,
13
+ "segment_size": 8192,
14
+ "init_lr_ratio": 1,
15
+ "warmup_epochs": 0,
16
+ "c_mel": 45,
17
+ "c_kl": 1.0
18
+ },
19
+ "data": {
20
+ "use_mel_posterior_encoder": true,
21
+ "training_files":"filelists/km_train.txt.cleaned",
22
+ "validation_files":"filelists/km_valid.txt.cleaned",
23
+ "text_cleaners":["canvers_km_cleaners"],
24
+ "max_wav_value": 32768.0,
25
+ "sampling_rate": 22050,
26
+ "filter_length": 1024,
27
+ "hop_length": 256,
28
+ "win_length": 1024,
29
+ "n_mel_channels": 80,
30
+ "mel_fmin": 0.0,
31
+ "mel_fmax": null,
32
+ "add_blank": false,
33
+ "n_speakers": 2,
34
+ "cleaned_text": true
35
+ },
36
+ "model": {
37
+ "use_mel_posterior_encoder": true,
38
+ "use_transformer_flows": true,
39
+ "transformer_flow_type": "pre_conv",
40
+ "use_spk_conditioned_encoder": true,
41
+ "use_noise_scaled_mas": true,
42
+ "use_duration_discriminator": true,
43
+ "inter_channels": 192,
44
+ "hidden_channels": 192,
45
+ "filter_channels": 768,
46
+ "n_heads": 2,
47
+ "n_layers": 6,
48
+ "kernel_size": 3,
49
+ "p_dropout": 0.1,
50
+ "resblock": "1",
51
+ "resblock_kernel_sizes": [3,7,11],
52
+ "resblock_dilation_sizes": [[1,3,5], [1,3,5], [1,3,5]],
53
+ "upsample_rates": [8,8,2,2],
54
+ "upsample_initial_channel": 512,
55
+ "upsample_kernel_sizes": [16,16,4,4],
56
+ "n_layers_q": 3,
57
+ "use_spectral_norm": false,
58
+ "use_sdp": false,
59
+ "gin_channels": 256
60
+ }
61
+ }
62
+
km_base.onnx ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:d2ff69b75153a332188c69c8651c996aa338f033242fd6461d10cb76fbf4146b
3
+ size 123427901
km_base.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:5fd276f892a72aac67f195c1f558b0bf7e438573f6850ce153ac608a4f6964b3
3
+ size 485792475
mn_base.json ADDED
@@ -0,0 +1,62 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "train": {
3
+ "log_interval": 1000,
4
+ "eval_interval": 5000,
5
+ "seed": 1234,
6
+ "epochs": 128,
7
+ "learning_rate": 1e-4,
8
+ "betas": [0.8, 0.99],
9
+ "eps": 1e-9,
10
+ "batch_size": 12,
11
+ "fp16_run": false,
12
+ "lr_decay": 0.999875,
13
+ "segment_size": 8192,
14
+ "init_lr_ratio": 1,
15
+ "warmup_epochs": 0,
16
+ "c_mel": 45,
17
+ "c_kl": 1.0
18
+ },
19
+ "data": {
20
+ "use_mel_posterior_encoder": true,
21
+ "training_files":"filelists/mn_train.txt.cleaned",
22
+ "validation_files":"filelists/mn_valid.txt.cleaned",
23
+ "text_cleaners":["canvers_mn_cleaners"],
24
+ "max_wav_value": 32768.0,
25
+ "sampling_rate": 22050,
26
+ "filter_length": 1024,
27
+ "hop_length": 256,
28
+ "win_length": 1024,
29
+ "n_mel_channels": 80,
30
+ "mel_fmin": 0.0,
31
+ "mel_fmax": null,
32
+ "add_blank": false,
33
+ "n_speakers": 2,
34
+ "cleaned_text": true
35
+ },
36
+ "model": {
37
+ "use_mel_posterior_encoder": true,
38
+ "use_transformer_flows": true,
39
+ "transformer_flow_type": "pre_conv",
40
+ "use_spk_conditioned_encoder": true,
41
+ "use_noise_scaled_mas": true,
42
+ "use_duration_discriminator": true,
43
+ "inter_channels": 192,
44
+ "hidden_channels": 192,
45
+ "filter_channels": 768,
46
+ "n_heads": 2,
47
+ "n_layers": 6,
48
+ "kernel_size": 3,
49
+ "p_dropout": 0.1,
50
+ "resblock": "1",
51
+ "resblock_kernel_sizes": [3,7,11],
52
+ "resblock_dilation_sizes": [[1,3,5], [1,3,5], [1,3,5]],
53
+ "upsample_rates": [8,8,2,2],
54
+ "upsample_initial_channel": 512,
55
+ "upsample_kernel_sizes": [16,16,4,4],
56
+ "n_layers_q": 3,
57
+ "use_spectral_norm": false,
58
+ "use_sdp": false,
59
+ "gin_channels": 256
60
+ }
61
+ }
62
+
mn_base.onnx ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:b9dcdd3ea993f46bd125e646823b9bdd2eedf7837fc037805a545c8e248c4664
3
+ size 123427901
mn_base.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:9f7e28579854c8fec23b1950b51484f3c3e12777cd58116f46d6d82da15e2a2c
3
+ size 485792475
tr_base.json ADDED
@@ -0,0 +1,62 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "train": {
3
+ "log_interval": 1000,
4
+ "eval_interval": 5000,
5
+ "seed": 1234,
6
+ "epochs": 128,
7
+ "learning_rate": 1e-4,
8
+ "betas": [0.8, 0.99],
9
+ "eps": 1e-9,
10
+ "batch_size": 12,
11
+ "fp16_run": false,
12
+ "lr_decay": 0.999875,
13
+ "segment_size": 8192,
14
+ "init_lr_ratio": 1,
15
+ "warmup_epochs": 0,
16
+ "c_mel": 45,
17
+ "c_kl": 1.0
18
+ },
19
+ "data": {
20
+ "use_mel_posterior_encoder": true,
21
+ "training_files":"filelists/tr_train.txt.cleaned",
22
+ "validation_files":"filelists/tr_valid.txt.cleaned",
23
+ "text_cleaners":["canvers_tr_cleaners"],
24
+ "max_wav_value": 32768.0,
25
+ "sampling_rate": 22050,
26
+ "filter_length": 1024,
27
+ "hop_length": 256,
28
+ "win_length": 1024,
29
+ "n_mel_channels": 80,
30
+ "mel_fmin": 0.0,
31
+ "mel_fmax": null,
32
+ "add_blank": false,
33
+ "n_speakers": 2,
34
+ "cleaned_text": true
35
+ },
36
+ "model": {
37
+ "use_mel_posterior_encoder": true,
38
+ "use_transformer_flows": true,
39
+ "transformer_flow_type": "pre_conv",
40
+ "use_spk_conditioned_encoder": true,
41
+ "use_noise_scaled_mas": true,
42
+ "use_duration_discriminator": true,
43
+ "inter_channels": 192,
44
+ "hidden_channels": 192,
45
+ "filter_channels": 768,
46
+ "n_heads": 2,
47
+ "n_layers": 6,
48
+ "kernel_size": 3,
49
+ "p_dropout": 0.1,
50
+ "resblock": "1",
51
+ "resblock_kernel_sizes": [3,7,11],
52
+ "resblock_dilation_sizes": [[1,3,5], [1,3,5], [1,3,5]],
53
+ "upsample_rates": [8,8,2,2],
54
+ "upsample_initial_channel": 512,
55
+ "upsample_kernel_sizes": [16,16,4,4],
56
+ "n_layers_q": 3,
57
+ "use_spectral_norm": false,
58
+ "use_sdp": false,
59
+ "gin_channels": 256
60
+ }
61
+ }
62
+
tr_base.onnx ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:6e4d97540abd053924e4865826ebcc0207a3ba13c337cb47d5478ae069a8377d
3
+ size 123427901
tr_base.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:4b02f0360abbf9b6a31a017bb849f7f7b51e1ff0fb86d1326d48315e065cb299
3
+ size 485792475