Hugging Face's logo

sl207
/
speecht5_anivoice_cloner

TensorBoard
Safetensors
speecht5
Model card Files
xet
Metrics Community
sl207 commited on
Commit
08ae468
·
verified ·
1 Parent(s): c956e48

Training in progress, step 100

Browse files
Files changed (26) hide show
  1. added_tokens.json +4 -0
  2. config.json +91 -0
  3. model.safetensors +3 -0
  4. preprocessor_config.json +19 -0
  5. runs/Aug18_17-43-08_1c0f66d5ba50/events.out.tfevents.1755539424.1c0f66d5ba50.226.0 +3 -0
  6. runs/Aug18_17-43-08_1c0f66d5ba50/events.out.tfevents.1755539559.1c0f66d5ba50.226.1 +3 -0
  7. runs/Aug18_17-43-08_1c0f66d5ba50/events.out.tfevents.1755540363.1c0f66d5ba50.226.2 +3 -0
  8. runs/Aug18_17-43-08_1c0f66d5ba50/events.out.tfevents.1755540469.1c0f66d5ba50.226.3 +3 -0
  9. runs/Aug18_17-43-08_1c0f66d5ba50/events.out.tfevents.1755541090.1c0f66d5ba50.226.4 +3 -0
  10. runs/Aug18_17-43-08_1c0f66d5ba50/events.out.tfevents.1755542099.1c0f66d5ba50.226.5 +3 -0
  11. runs/Aug18_17-43-08_1c0f66d5ba50/events.out.tfevents.1755542138.1c0f66d5ba50.226.6 +3 -0
  12. runs/Aug18_17-43-08_1c0f66d5ba50/events.out.tfevents.1755542184.1c0f66d5ba50.226.7 +3 -0
  13. runs/Aug18_18-39-42_1c0f66d5ba50/events.out.tfevents.1755542389.1c0f66d5ba50.226.8 +3 -0
  14. runs/Aug18_18-39-42_1c0f66d5ba50/events.out.tfevents.1755542652.1c0f66d5ba50.226.9 +3 -0
  15. runs/Aug18_18-39-42_1c0f66d5ba50/events.out.tfevents.1755542717.1c0f66d5ba50.226.10 +3 -0
  16. runs/Aug18_18-39-42_1c0f66d5ba50/events.out.tfevents.1755542938.1c0f66d5ba50.226.11 +3 -0
  17. runs/Aug18_18-53-08_1c0f66d5ba50/events.out.tfevents.1755543194.1c0f66d5ba50.226.12 +3 -0
  18. runs/Aug18_18-53-08_1c0f66d5ba50/events.out.tfevents.1755543309.1c0f66d5ba50.226.13 +3 -0
  19. runs/Aug18_18-53-08_1c0f66d5ba50/events.out.tfevents.1755543750.1c0f66d5ba50.226.14 +3 -0
  20. runs/Aug18_18-53-08_1c0f66d5ba50/events.out.tfevents.1755544175.1c0f66d5ba50.226.15 +3 -0
  21. runs/Aug18_18-53-08_1c0f66d5ba50/events.out.tfevents.1755544410.1c0f66d5ba50.226.16 +3 -0
  22. runs/Aug19_14-04-30_572c756e01aa/events.out.tfevents.1755612359.572c756e01aa.187.0 +3 -0
  23. special_tokens_map.json +13 -0
  24. spm_char.model +3 -0
  25. tokenizer_config.json +64 -0
  26. training_args.bin +3 -0
added_tokens.json ADDED
@@ -0,0 +1,4 @@
 
 
 
 
 
1
+ {
2
+ "<ctc_blank>": 80,
3
+ "<mask>": 79
4
+ }
config.json ADDED
@@ -0,0 +1,91 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "activation_dropout": 0.1,
3
+ "apply_spec_augment": true,
4
+ "architectures": [
5
+ "SpeechT5ForTextToSpeech"
6
+ ],
7
+ "attention_dropout": 0.1,
8
+ "bos_token_id": 0,
9
+ "conv_bias": false,
10
+ "conv_dim": [
11
+ 512,
12
+ 512,
13
+ 512,
14
+ 512,
15
+ 512,
16
+ 512,
17
+ 512
18
+ ],
19
+ "conv_kernel": [
20
+ 10,
21
+ 3,
22
+ 3,
23
+ 3,
24
+ 3,
25
+ 2,
26
+ 2
27
+ ],
28
+ "conv_stride": [
29
+ 5,
30
+ 2,
31
+ 2,
32
+ 2,
33
+ 2,
34
+ 2,
35
+ 2
36
+ ],
37
+ "decoder_attention_heads": 12,
38
+ "decoder_ffn_dim": 3072,
39
+ "decoder_layerdrop": 0.1,
40
+ "decoder_layers": 6,
41
+ "decoder_start_token_id": 2,
42
+ "encoder_attention_heads": 12,
43
+ "encoder_ffn_dim": 3072,
44
+ "encoder_layerdrop": 0.1,
45
+ "encoder_layers": 12,
46
+ "encoder_max_relative_position": 160,
47
+ "eos_token_id": 2,
48
+ "feat_extract_activation": "gelu",
49
+ "feat_extract_norm": "group",
50
+ "feat_proj_dropout": 0.0,
51
+ "guided_attention_loss_num_heads": 2,
52
+ "guided_attention_loss_scale": 10.0,
53
+ "guided_attention_loss_sigma": 0.4,
54
+ "hidden_act": "gelu",
55
+ "hidden_dropout": 0.1,
56
+ "hidden_size": 768,
57
+ "initializer_range": 0.02,
58
+ "is_encoder_decoder": true,
59
+ "layer_norm_eps": 1e-05,
60
+ "mask_feature_length": 10,
61
+ "mask_feature_min_masks": 0,
62
+ "mask_feature_prob": 0.0,
63
+ "mask_time_length": 10,
64
+ "mask_time_min_masks": 2,
65
+ "mask_time_prob": 0.05,
66
+ "max_length": null,
67
+ "max_speech_positions": 1876,
68
+ "max_text_positions": 600,
69
+ "model_type": "speecht5",
70
+ "num_conv_pos_embedding_groups": 16,
71
+ "num_conv_pos_embeddings": 128,
72
+ "num_feat_extract_layers": 7,
73
+ "num_mel_bins": 80,
74
+ "pad_token_id": 1,
75
+ "positional_dropout": 0.1,
76
+ "reduction_factor": 2,
77
+ "scale_embedding": false,
78
+ "speaker_embedding_dim": 512,
79
+ "speech_decoder_postnet_dropout": 0.5,
80
+ "speech_decoder_postnet_kernel": 5,
81
+ "speech_decoder_postnet_layers": 5,
82
+ "speech_decoder_postnet_units": 256,
83
+ "speech_decoder_prenet_dropout": 0.5,
84
+ "speech_decoder_prenet_layers": 2,
85
+ "speech_decoder_prenet_units": 256,
86
+ "torch_dtype": "float32",
87
+ "transformers_version": "4.52.4",
88
+ "use_cache": false,
89
+ "use_guided_attention_loss": true,
90
+ "vocab_size": 81
91
+ }
model.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:5ffde0499a750f0c00e6202dcb3f392c1486eb577f4b8e3361763e1c90a39772
3
+ size 577789320
preprocessor_config.json ADDED
@@ -0,0 +1,19 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "do_normalize": false,
3
+ "feature_extractor_type": "SpeechT5FeatureExtractor",
4
+ "feature_size": 1,
5
+ "fmax": 7600,
6
+ "fmin": 80,
7
+ "frame_signal_scale": 1.0,
8
+ "hop_length": 16,
9
+ "mel_floor": 1e-10,
10
+ "num_mel_bins": 80,
11
+ "padding_side": "right",
12
+ "padding_value": 0.0,
13
+ "processor_class": "SpeechT5Processor",
14
+ "reduction_factor": 2,
15
+ "return_attention_mask": true,
16
+ "sampling_rate": 16000,
17
+ "win_function": "hann_window",
18
+ "win_length": 64
19
+ }
runs/Aug18_17-43-08_1c0f66d5ba50/events.out.tfevents.1755539424.1c0f66d5ba50.226.0 ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:a34c2559e7bc0fb3df16e4cf9c30c72e388f2183b98f2f9b94a8189555a20915
3
+ size 19408
runs/Aug18_17-43-08_1c0f66d5ba50/events.out.tfevents.1755539559.1c0f66d5ba50.226.1 ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:490250720dcabacb5caa39b3e6af1020eb63e8cd0ff5d4143c4431ab11e6997e
3
+ size 6528
runs/Aug18_17-43-08_1c0f66d5ba50/events.out.tfevents.1755540363.1c0f66d5ba50.226.2 ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:8e2cbe7b6ae4bf670c47f2e50580a772a43fd8b7f897f5c00ccbe0fedf07f808
3
+ size 6528
runs/Aug18_17-43-08_1c0f66d5ba50/events.out.tfevents.1755540469.1c0f66d5ba50.226.3 ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:8854511f63e741c95e1592347c300d2516bcb5e67e5096273c50da1b06f37a87
3
+ size 6528
runs/Aug18_17-43-08_1c0f66d5ba50/events.out.tfevents.1755541090.1c0f66d5ba50.226.4 ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:72d30974e36f54d4d3ec1cbe48cd55ce3cc35549b8db42854865eec5138bf375
3
+ size 12968
runs/Aug18_17-43-08_1c0f66d5ba50/events.out.tfevents.1755542099.1c0f66d5ba50.226.5 ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:6056e957239e31788997b26324af467d2a4620ff1d5e75139e6b274367d41db6
3
+ size 6528
runs/Aug18_17-43-08_1c0f66d5ba50/events.out.tfevents.1755542138.1c0f66d5ba50.226.6 ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:97b37f53cfcf421d254a59e4d2b86fe49b34a5879d3d791a54a9f0816486ed1c
3
+ size 6528
runs/Aug18_17-43-08_1c0f66d5ba50/events.out.tfevents.1755542184.1c0f66d5ba50.226.7 ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:cb002900b38bf721a523a2606e0f9a85e0731889c365420e6e27f8661d6e2b6a
3
+ size 6528
runs/Aug18_18-39-42_1c0f66d5ba50/events.out.tfevents.1755542389.1c0f66d5ba50.226.8 ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:5eca7da947fb80fbd0ca3964108a2f79ffa5005484a13830c8904384d61e0cc1
3
+ size 6528
runs/Aug18_18-39-42_1c0f66d5ba50/events.out.tfevents.1755542652.1c0f66d5ba50.226.9 ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:f9abd319767235886feaeb487e84e7757a0a5421c5fd388195e68905360ae2b0
3
+ size 6528
runs/Aug18_18-39-42_1c0f66d5ba50/events.out.tfevents.1755542717.1c0f66d5ba50.226.10 ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:d853dfc4f9e5e5073fa3fb69a56ce2708e23e0c78f8edee8299461c7c7ace49c
3
+ size 6528
runs/Aug18_18-39-42_1c0f66d5ba50/events.out.tfevents.1755542938.1c0f66d5ba50.226.11 ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:130578744a2ac81bd9f0408a9a39bfee5c59b37e5d8efab41086ba80324e4072
3
+ size 6528
runs/Aug18_18-53-08_1c0f66d5ba50/events.out.tfevents.1755543194.1c0f66d5ba50.226.12 ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:133a6a48281ae5a30ac145f3c9aa714c9eddb8a16abfb9c088991c3227e676a2
3
+ size 6528
runs/Aug18_18-53-08_1c0f66d5ba50/events.out.tfevents.1755543309.1c0f66d5ba50.226.13 ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:878b9cbb1354521d228afd72473629dd5cf18758588e48e3306ff4915d491c50
3
+ size 6528
runs/Aug18_18-53-08_1c0f66d5ba50/events.out.tfevents.1755543750.1c0f66d5ba50.226.14 ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:5a40fdce0e6072da179a5ed0decda6a2d2a08180956992591b910e2d535aaa4e
3
+ size 6528
runs/Aug18_18-53-08_1c0f66d5ba50/events.out.tfevents.1755544175.1c0f66d5ba50.226.15 ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:d172f2106be77cc04f893e1c46b59a19ecf5a1469114e419b23e010cbc5a3650
3
+ size 6528
runs/Aug18_18-53-08_1c0f66d5ba50/events.out.tfevents.1755544410.1c0f66d5ba50.226.16 ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:5dd84a2596f56f2198626fcd584509bef2e74ba98540452642250a553a234166
3
+ size 4184
runs/Aug19_14-04-30_572c756e01aa/events.out.tfevents.1755612359.572c756e01aa.187.0 ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:665ae1d772f9bb38454f92da4bf169fa0d6372af167d8f3e533f8233f6995074
3
+ size 6541
special_tokens_map.json ADDED
@@ -0,0 +1,13 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "bos_token": "<s>",
3
+ "eos_token": "</s>",
4
+ "mask_token": {
5
+ "content": "<mask>",
6
+ "lstrip": true,
7
+ "normalized": true,
8
+ "rstrip": false,
9
+ "single_word": false
10
+ },
11
+ "pad_token": "<pad>",
12
+ "unk_token": "<unk>"
13
+ }
spm_char.model ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:7fcc48f3e225f627b1641db410ceb0c8649bd2b0c982e150b03f8be3728ab560
3
+ size 238473
tokenizer_config.json ADDED
@@ -0,0 +1,64 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "added_tokens_decoder": {
3
+ "0": {
4
+ "content": "<s>",
5
+ "lstrip": false,
6
+ "normalized": false,
7
+ "rstrip": false,
8
+ "single_word": false,
9
+ "special": true
10
+ },
11
+ "1": {
12
+ "content": "<pad>",
13
+ "lstrip": false,
14
+ "normalized": false,
15
+ "rstrip": false,
16
+ "single_word": false,
17
+ "special": true
18
+ },
19
+ "2": {
20
+ "content": "</s>",
21
+ "lstrip": false,
22
+ "normalized": false,
23
+ "rstrip": false,
24
+ "single_word": false,
25
+ "special": true
26
+ },
27
+ "3": {
28
+ "content": "<unk>",
29
+ "lstrip": false,
30
+ "normalized": false,
31
+ "rstrip": false,
32
+ "single_word": false,
33
+ "special": true
34
+ },
35
+ "79": {
36
+ "content": "<mask>",
37
+ "lstrip": true,
38
+ "normalized": true,
39
+ "rstrip": false,
40
+ "single_word": false,
41
+ "special": true
42
+ },
43
+ "80": {
44
+ "content": "<ctc_blank>",
45
+ "lstrip": false,
46
+ "normalized": true,
47
+ "rstrip": false,
48
+ "single_word": false,
49
+ "special": false
50
+ }
51
+ },
52
+ "bos_token": "<s>",
53
+ "clean_up_tokenization_spaces": false,
54
+ "eos_token": "</s>",
55
+ "extra_special_tokens": {},
56
+ "mask_token": "<mask>",
57
+ "model_max_length": 600,
58
+ "normalize": false,
59
+ "pad_token": "<pad>",
60
+ "processor_class": "SpeechT5Processor",
61
+ "sp_model_kwargs": {},
62
+ "tokenizer_class": "SpeechT5Tokenizer",
63
+ "unk_token": "<unk>"
64
+ }
training_args.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:6ceeafe361b99656b77b85c4a29cb9c44468295db664be29d8c748f4a1a8d2f1
3
+ size 5432