Upload 9 files
Browse files- ar_base.json +62 -0
- ar_base.onnx +3 -0
- ar_base.pth +3 -0
- fr_base.json +62 -0
- fr_base.onnx +3 -0
- fr_base.pth +3 -0
- pt_base.json +62 -0
- pt_base.onnx +3 -0
- pt_base.pth +3 -0
ar_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/ar_train.txt.cleaned",
|
22 |
+
"validation_files":"filelists/ar_valid.txt.cleaned",
|
23 |
+
"text_cleaners":["canvers_ar_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": 32,
|
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 |
+
|
ar_base.onnx
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:90bdd7041eb1e05d1d354c060530ece9b0883bbff50de069b15a9954273511f3
|
3 |
+
size 123458623
|
ar_base.pth
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:b1902af832b8263a84059f69175c2c709e95e2bf8b1281d27bfeabb77aac59b7
|
3 |
+
size 485887621
|
fr_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": 6,
|
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/fr_train.txt.cleaned",
|
22 |
+
"validation_files":"filelists/fr_valid.txt.cleaned",
|
23 |
+
"text_cleaners":["canvers_fr_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": 21,
|
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 |
+
|
fr_base.onnx
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:940e854dd848986f5833cc2efc68825132b5809993f29b14e484135f7e4918b7
|
3 |
+
size 123447359
|
fr_base.pth
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:e72aff2772ebd997b7e986f36fed91dec4c8216f86c781a6347c012a9528659f
|
3 |
+
size 485853829
|
pt_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/pt_train.txt.cleaned",
|
22 |
+
"validation_files":"filelists/pt_valid.txt.cleaned",
|
23 |
+
"text_cleaners":["canvers_pt_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": 19,
|
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 |
+
|
pt_base.onnx
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:cc8735e66e47978a1541f22cc97b33fa0009f94b777eab30992bc7f4a15b0c90
|
3 |
+
size 123445311
|
pt_base.pth
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:afc1d93bfd700f0e72260d6abf1f5dc02c7a1f5b4e77532ecebd5b1877f91f21
|
3 |
+
size 485844699
|