Hugging Face's logo

ForHHeart
/
speecht5_finetuned_sawahili

Text-to-Audio
Transformers
Safetensors
speecht5
Inference Endpoints
Model card Files Community
ForHHeart commited on
Commit
60de872
•
1 Parent(s): ccb7f5d

Upload SpeechT5ForTextToSpeech

Browse files
Files changed (2) hide show
  1. config.json +83 -40
  2. model.safetensors +2 -2
config.json CHANGED
@@ -1,49 +1,92 @@
1
  {
2
- "_name_or_path": "microsoft/speecht5_hifigan",
 
 
3
  "architectures": [
4
- "SpeechT5HifiGan"
5
  ],
6
- "initializer_range": 0.01,
7
- "leaky_relu_slope": 0.1,
8
- "model_in_dim": 80,
9
- "model_type": "hifigan",
10
- "normalize_before": true,
11
- "resblock_dilation_sizes": [
12
- [
13
- 1,
14
- 3,
15
- 5
16
- ],
17
- [
18
- 1,
19
- 3,
20
- 5
21
- ],
22
- [
23
- 1,
24
- 3,
25
- 5
26
- ]
27
  ],
28
- "resblock_kernel_sizes": [
 
29
  3,
30
- 7,
31
- 11
 
 
 
32
  ],
33
- "sampling_rate": 16000,
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
34
  "torch_dtype": "float32",
35
  "transformers_version": "4.40.0",
36
- "upsample_initial_channel": 512,
37
- "upsample_kernel_sizes": [
38
- 8,
39
- 8,
40
- 8,
41
- 8
42
- ],
43
- "upsample_rates": [
44
- 4,
45
- 4,
46
- 4,
47
- 4
48
- ]
49
  }
 
1
  {
2
+ "_name_or_path": "microsoft/speecht5_tts",
3
+ "activation_dropout": 0.1,
4
+ "apply_spec_augment": true,
5
  "architectures": [
6
+ "SpeechT5ForTextToSpeech"
7
  ],
8
+ "attention_dropout": 0.1,
9
+ "bos_token_id": 0,
10
+ "conv_bias": false,
11
+ "conv_dim": [
12
+ 512,
13
+ 512,
14
+ 512,
15
+ 512,
16
+ 512,
17
+ 512,
18
+ 512
 
 
 
 
 
 
 
 
 
 
19
  ],
20
+ "conv_kernel": [
21
+ 10,
22
  3,
23
+ 3,
24
+ 3,
25
+ 3,
26
+ 2,
27
+ 2
28
  ],
29
+ "conv_stride": [
30
+ 5,
31
+ 2,
32
+ 2,
33
+ 2,
34
+ 2,
35
+ 2,
36
+ 2
37
+ ],
38
+ "decoder_attention_heads": 12,
39
+ "decoder_ffn_dim": 3072,
40
+ "decoder_layerdrop": 0.1,
41
+ "decoder_layers": 6,
42
+ "decoder_start_token_id": 2,
43
+ "encoder_attention_heads": 12,
44
+ "encoder_ffn_dim": 3072,
45
+ "encoder_layerdrop": 0.1,
46
+ "encoder_layers": 12,
47
+ "encoder_max_relative_position": 160,
48
+ "eos_token_id": 2,
49
+ "feat_extract_activation": "gelu",
50
+ "feat_extract_norm": "group",
51
+ "feat_proj_dropout": 0.0,
52
+ "guided_attention_loss_num_heads": 2,
53
+ "guided_attention_loss_scale": 10.0,
54
+ "guided_attention_loss_sigma": 0.4,
55
+ "hidden_act": "gelu",
56
+ "hidden_dropout": 0.1,
57
+ "hidden_size": 768,
58
+ "initializer_range": 0.02,
59
+ "is_encoder_decoder": true,
60
+ "layer_norm_eps": 1e-05,
61
+ "mask_feature_length": 10,
62
+ "mask_feature_min_masks": 0,
63
+ "mask_feature_prob": 0.0,
64
+ "mask_time_length": 10,
65
+ "mask_time_min_masks": 2,
66
+ "mask_time_prob": 0.05,
67
+ "max_length": 1876,
68
+ "max_speech_positions": 1876,
69
+ "max_text_positions": 600,
70
+ "model_type": "speecht5",
71
+ "num_conv_pos_embedding_groups": 16,
72
+ "num_conv_pos_embeddings": 128,
73
+ "num_feat_extract_layers": 7,
74
+ "num_mel_bins": 80,
75
+ "pad_token_id": 1,
76
+ "positional_dropout": 0.1,
77
+ "reduction_factor": 2,
78
+ "scale_embedding": false,
79
+ "speaker_embedding_dim": 512,
80
+ "speech_decoder_postnet_dropout": 0.5,
81
+ "speech_decoder_postnet_kernel": 5,
82
+ "speech_decoder_postnet_layers": 5,
83
+ "speech_decoder_postnet_units": 256,
84
+ "speech_decoder_prenet_dropout": 0.5,
85
+ "speech_decoder_prenet_layers": 2,
86
+ "speech_decoder_prenet_units": 256,
87
  "torch_dtype": "float32",
88
  "transformers_version": "4.40.0",
89
+ "use_cache": true,
90
+ "use_guided_attention_loss": true,
91
+ "vocab_size": 81
 
 
 
 
 
 
 
 
 
 
92
  }
model.safetensors CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:5e1ae998705b24b74b79b55bd9015f458ee85dbc1e98448aa7f0e0e066eba1bd
3
- size 50640724
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:17cb8139940fda044376037a5c59f130688e1fbd88589f2ac9ef58366609200b
3
+ size 577789320