subatomicseer commited on
Commit
5cd0a6b
1 Parent(s): 3f52902

Upload 13 files

Browse files
.gitattributes CHANGED
@@ -33,3 +33,4 @@ saved_model/**/* filter=lfs diff=lfs merge=lfs -text
33
  *.zip filter=lfs diff=lfs merge=lfs -text
34
  *.zst filter=lfs diff=lfs merge=lfs -text
35
  *tfevents* filter=lfs diff=lfs merge=lfs -text
 
 
33
  *.zip filter=lfs diff=lfs merge=lfs -text
34
  *.zst filter=lfs diff=lfs merge=lfs -text
35
  *tfevents* filter=lfs diff=lfs merge=lfs -text
36
+ denoiser/g_best filter=lfs diff=lfs merge=lfs -text
denoiser/g_best ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:0088cd06068a97b97cc13fe10fe155ea5c24beea79564b2162fab22a79dc9dc5
3
+ size 8350488
hierspeechpp_eng_kor/config.json ADDED
@@ -0,0 +1,67 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "train": {
3
+ "log_interval": 200,
4
+ "eval_interval": 10000,
5
+ "save_interval": 10000,
6
+ "seed": 1234,
7
+ "epochs": 20000,
8
+ "learning_rate": 1e-4,
9
+ "betas": [0.8, 0.99],
10
+ "eps": 1e-9,
11
+ "batch_size": 20,
12
+ "fp16_run": false,
13
+ "lr_decay": 0.999,
14
+ "segment_size": 61440,
15
+ "init_lr_ratio": 1,
16
+ "warmup_epochs": 0,
17
+ "c_mel": 45,
18
+ "c_kl": 1.0,
19
+ "c_bi_kl": 0.5,
20
+ "c_mixup": 1,
21
+ "c_pho": 45.0,
22
+ "c_f0": 45,
23
+ "aug": true
24
+ },
25
+ "data": {
26
+ "train_filelist_path": "filelists/train_eng_kor_wav.txt",
27
+ "test_filelist_path": "filelists/test_hier_16k_72_wav.txt",
28
+ "text_cleaners":["english_cleaners2"],
29
+ "max_wav_value": 32768.0,
30
+ "sampling_rate": 16000,
31
+ "filter_length": 1280,
32
+ "hop_length": 320,
33
+ "win_length": 1280,
34
+ "n_mel_channels": 80,
35
+ "mel_fmin": 0,
36
+ "mel_fmax": 8000,
37
+ "add_blank": true,
38
+ "n_speakers": 0,
39
+ "cleaned_text": true,
40
+ "aug_rate": 1.0,
41
+ "top_db": 20
42
+ },
43
+ "model": {
44
+ "inter_channels": 192,
45
+ "hidden_channels": 192,
46
+ "filter_channels": 768,
47
+ "n_heads": 2,
48
+ "n_layers": 6,
49
+ "kernel_size": 3,
50
+ "p_dropout": 0.1,
51
+ "resblock": "1",
52
+ "resblock_kernel_sizes": [3,7,11],
53
+ "resblock_dilation_sizes": [[1,3,5], [1,3,5], [1,3,5]],
54
+ "upsample_rates": [4,5,4,2,2],
55
+ "upsample_initial_channel": 512,
56
+ "upsample_kernel_sizes": [8,11,8,4,4],
57
+ "n_layers_q": 3,
58
+ "use_spectral_norm": false,
59
+ "wav2vec_feature_layer": 7,
60
+ "gin_channels": 256,
61
+ "cfg": true,
62
+ "uncond_ratio": 0.1,
63
+ "prosody_size": 20
64
+
65
+ }
66
+ }
67
+
hierspeechpp_eng_kor/hierspeechpp_v1.1_ckpt.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:9fd704f9a4d11b62195caaaeb76ce759b12ea69a1c36b2716af44bd60bde6a64
3
+ size 388999163
hierspeechpp_eng_kor/hierspeechpp_v1_ckpt.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:0229501cd273135bce9ee0be09ddef18e079cb8dbe61cda665fec73afbfad0d4
3
+ size 388970075
hierspeechpp_libritts460/config.json ADDED
@@ -0,0 +1,67 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "train": {
3
+ "log_interval": 200,
4
+ "eval_interval": 10000,
5
+ "save_interval": 10000,
6
+ "seed": 1234,
7
+ "epochs": 20000,
8
+ "learning_rate": 1e-4,
9
+ "betas": [0.8, 0.99],
10
+ "eps": 1e-9,
11
+ "batch_size": 16,
12
+ "fp16_run": false,
13
+ "lr_decay": 0.999875,
14
+ "segment_size": 61440,
15
+ "init_lr_ratio": 1,
16
+ "warmup_epochs": 0,
17
+ "c_mel": 45,
18
+ "c_kl": 1.0,
19
+ "c_bi_kl": 0.5,
20
+ "c_mixup": 1,
21
+ "c_pho": 45.0,
22
+ "c_f0": 45,
23
+ "aug": true
24
+ },
25
+ "data": {
26
+ "train_filelist_path": "filelists/train_libritts_460_wav.txt",
27
+ "test_filelist_path": "filelists/test_hier_16k_72_wav.txt",
28
+ "text_cleaners":["english_cleaners2"],
29
+ "max_wav_value": 32768.0,
30
+ "sampling_rate": 16000,
31
+ "filter_length": 1280,
32
+ "hop_length": 320,
33
+ "win_length": 1280,
34
+ "n_mel_channels": 80,
35
+ "mel_fmin": 0,
36
+ "mel_fmax": 8000,
37
+ "add_blank": true,
38
+ "n_speakers": 0,
39
+ "cleaned_text": true,
40
+ "aug_rate": 1.0,
41
+ "top_db": 20
42
+ },
43
+ "model": {
44
+ "inter_channels": 192,
45
+ "hidden_channels": 192,
46
+ "filter_channels": 768,
47
+ "n_heads": 2,
48
+ "n_layers": 6,
49
+ "kernel_size": 3,
50
+ "p_dropout": 0.1,
51
+ "resblock": "1",
52
+ "resblock_kernel_sizes": [3,7,11],
53
+ "resblock_dilation_sizes": [[1,3,5], [1,3,5], [1,3,5]],
54
+ "upsample_rates": [4,5,4,2,2],
55
+ "upsample_initial_channel": 512,
56
+ "upsample_kernel_sizes": [8,11,8,4,4],
57
+ "n_layers_q": 3,
58
+ "use_spectral_norm": false,
59
+ "wav2vec_feature_layer": 7,
60
+ "gin_channels": 256,
61
+ "cfg": true,
62
+ "uncond_ratio": 0.1,
63
+ "prosody_size": 20
64
+
65
+ }
66
+ }
67
+
hierspeechpp_libritts460/hierspeechpp_lt460_ckpt.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:9f6928ad968454061b30fbeb024a4bbd9bfcadfd5e2660c213b95ab83fa3f171
3
+ size 389002475
hierspeechpp_libritts960/config.json ADDED
@@ -0,0 +1,67 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "train": {
3
+ "log_interval": 200,
4
+ "eval_interval": 10000,
5
+ "save_interval": 10000,
6
+ "seed": 1234,
7
+ "epochs": 20000,
8
+ "learning_rate": 1e-4,
9
+ "betas": [0.8, 0.99],
10
+ "eps": 1e-9,
11
+ "batch_size": 16,
12
+ "fp16_run": false,
13
+ "lr_decay": 0.999875,
14
+ "segment_size": 61440,
15
+ "init_lr_ratio": 1,
16
+ "warmup_epochs": 0,
17
+ "c_mel": 45,
18
+ "c_kl": 1.0,
19
+ "c_bi_kl": 0.5,
20
+ "c_mixup": 1,
21
+ "c_pho": 45.0,
22
+ "c_f0": 45,
23
+ "aug": true
24
+ },
25
+ "data": {
26
+ "train_filelist_path": "filelists/train_libritts_all_wav.txt",
27
+ "test_filelist_path": "filelists/test_libritts_all_wav.txt",
28
+ "text_cleaners":["english_cleaners2"],
29
+ "max_wav_value": 32768.0,
30
+ "sampling_rate": 16000,
31
+ "filter_length": 1280,
32
+ "hop_length": 320,
33
+ "win_length": 1280,
34
+ "n_mel_channels": 80,
35
+ "mel_fmin": 0,
36
+ "mel_fmax": 8000,
37
+ "add_blank": true,
38
+ "n_speakers": 0,
39
+ "cleaned_text": true,
40
+ "aug_rate": 1.0,
41
+ "top_db": 20
42
+ },
43
+ "model": {
44
+ "inter_channels": 192,
45
+ "hidden_channels": 192,
46
+ "filter_channels": 768,
47
+ "n_heads": 2,
48
+ "n_layers": 6,
49
+ "kernel_size": 3,
50
+ "p_dropout": 0.1,
51
+ "resblock": "1",
52
+ "resblock_kernel_sizes": [3,7,11],
53
+ "resblock_dilation_sizes": [[1,3,5], [1,3,5], [1,3,5]],
54
+ "upsample_rates": [4,5,4,2,2],
55
+ "upsample_initial_channel": 512,
56
+ "upsample_kernel_sizes": [8,11,8,4,4],
57
+ "n_layers_q": 3,
58
+ "use_spectral_norm": false,
59
+ "wav2vec_feature_layer": 7,
60
+ "gin_channels": 256,
61
+ "cfg": true,
62
+ "uncond_ratio": 0.1,
63
+ "prosody_size": 20
64
+
65
+ }
66
+ }
67
+
hierspeechpp_libritts960/hierspeechpp_lt960_ckpt.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:609465feba7cee4a1dd8e25e456459cb5ca0f4cb00d97fb1f7007e422d1a6fd6
3
+ size 389002475
speechsr24k/G_340000.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:94bc84b6f8bdb375e2027bc7a3222730d29f9d4c042fd9edce0024806d5c4320
3
+ size 1715101
speechsr48k/G_100000.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:62c70874ac4efeb4dc9c8aa9dc0a611a951e1c36292abeb4c406d7fb91e0eefc
3
+ size 1715101
ttv_libritts_v1/G_950000.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:0b0fdd441b010432c8e10903195980be6c94748491050b9e723a9e180c3b1e81
3
+ size 1303530971
ttv_libritts_v1/config.json ADDED
@@ -0,0 +1,55 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "train": {
3
+ "log_interval": 200,
4
+ "eval_interval": 10000,
5
+ "save_interval": 10000,
6
+ "seed": 1234,
7
+ "epochs": 20000,
8
+ "learning_rate": 2e-4,
9
+ "betas": [0.8, 0.99],
10
+ "eps": 1e-9,
11
+ "batch_size": 32,
12
+ "fp16_run": false,
13
+ "lr_decay": 0.999,
14
+ "segment_size": 192000,
15
+ "init_lr_ratio": 1,
16
+ "warmup_epochs": 0,
17
+ "c_mel": 10,
18
+ "c_kl": 0.1,
19
+ "c_pho": 45.0,
20
+ "c_f0": 1,
21
+ "aug": true
22
+ },
23
+ "data": {
24
+ "train_filelist_path": "filelists/train_wav.txt",
25
+ "test_filelist_path": "filelists/test_wav.txt",
26
+ "text_cleaners":["english_cleaners2"],
27
+ "max_wav_value": 32768.0,
28
+ "sampling_rate": 16000,
29
+ "filter_length": 1280,
30
+ "hop_length": 320,
31
+ "win_length": 1280,
32
+ "n_mel_channels": 80,
33
+ "mel_fmin": 0,
34
+ "mel_fmax": 8000,
35
+ "add_blank": true,
36
+ "n_speakers": 0,
37
+ "cleaned_text": true,
38
+ "aug_rate": 1.0,
39
+ "top_db": 20
40
+ },
41
+ "model": {
42
+ "inter_channels": 256,
43
+ "hidden_channels": 256,
44
+ "filter_channels": 1024,
45
+ "n_heads": 4,
46
+ "n_layers": 6,
47
+ "kernel_size": 3,
48
+ "p_dropout": 0.1,
49
+ "resblock": "1",
50
+ "resblock_kernel_sizes": [3,7,11],
51
+ "resblock_dilation_sizes": [[1,3,5], [1,3,5], [1,3,5]],
52
+ "use_spectral_norm": false
53
+ }
54
+ }
55
+
ttv_libritts_v1/ttv_lt960_ckpt.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:2e408e2c6a72551da0ca616f43e0f989449ff5e0d0ca6229b27dea3dc36aa6c1
3
+ size 434482782