Hugging Face's logo

Xenova
/
speecht5_tts

Text-to-Speech
Transformers.js
ONNX
speecht5
text-to-audio
Model card Files Community
4
Xenova HF staff commited on
Commit
ec6335c
•
1 Parent(s): 2b26158

Upload folder using huggingface_hub

Browse files
Files changed (12) hide show
  1. added_tokens.json +1 -5
  2. config.json +1 -1
  3. generation_config.json +1 -1
  4. onnx/decoder_model.onnx +2 -2
  5. onnx/decoder_model_merged.onnx +3 -0
  6. onnx/decoder_model_merged_quantized.onnx +3 -0
  7. onnx/decoder_model_quantized.onnx +2 -2
  8. onnx/decoder_with_past_model.onnx +3 -0
  9. onnx/decoder_with_past_model_quantized.onnx +3 -0
  10. quantize_config.json +116 -47
  11. special_tokens_map.json +7 -0
  12. tokenizer_config.json +0 -52
added_tokens.json CHANGED
@@ -1,8 +1,4 @@
1
  {
2
- "</s>": 2,
3
  "<ctc_blank>": 80,
4
- "<mask>": 79,
5
- "<pad>": 1,
6
- "<s>": 0,
7
- "<unk>": 3
8
  }
 
1
  {
 
2
  "<ctc_blank>": 80,
3
+ "<mask>": 79
 
 
 
4
  }
config.json CHANGED
@@ -84,7 +84,7 @@
84
  "speech_decoder_prenet_dropout": 0.5,
85
  "speech_decoder_prenet_layers": 2,
86
  "speech_decoder_prenet_units": 256,
87
- "transformers_version": "4.34.0.dev0",
88
  "use_cache": true,
89
  "use_guided_attention_loss": true,
90
  "vocab_size": 81
 
84
  "speech_decoder_prenet_dropout": 0.5,
85
  "speech_decoder_prenet_layers": 2,
86
  "speech_decoder_prenet_units": 256,
87
+ "transformers_version": "4.33.2",
88
  "use_cache": true,
89
  "use_guided_attention_loss": true,
90
  "vocab_size": 81
generation_config.json CHANGED
@@ -5,5 +5,5 @@
5
  "eos_token_id": 2,
6
  "max_length": 1876,
7
  "pad_token_id": 1,
8
- "transformers_version": "4.34.0.dev0"
9
  }
 
5
  "eos_token_id": 2,
6
  "max_length": 1876,
7
  "pad_token_id": 1,
8
+ "transformers_version": "4.33.2"
9
  }
onnx/decoder_model.onnx CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:f8f9a2eaf23c51181f115dfde708c8caf9e22b18fd148b680bcea59da58977fe
3
- size 238455017
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:11b6df19b8b0ec7b847d2e5407c9f9f4cd0feaa6a429aafe2c8621d85ecd15b4
3
+ size 238455269
onnx/decoder_model_merged.onnx ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:62b9794ba40b31a96226f5052a594bc3c4cd24f8f523e16607b4402963c7e3cf
3
+ size 244478567
onnx/decoder_model_merged_quantized.onnx ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:7130cc76c2fdd3c833e59024ad741cc3797460b9fc86c542d95e86e4c04b203c
3
+ size 71075734
onnx/decoder_model_quantized.onnx CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:75879674ca1f08b6ca794bb3965ecc5392b1af329516457ba91794bbeb07b6d4
3
- size 64901194
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:ccf952066f11dc85efb025e3f3f085bcdf3f2d88206aec2e814bd2b4d4da8417
3
+ size 64899850
onnx/decoder_with_past_model.onnx ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:7da4d3fe8cd052d777a71c6704737ddf681f961fc222ff7c09b6719c516ec1f1
3
+ size 210086828
onnx/decoder_with_past_model_quantized.onnx ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:eb1036eebb535917ef0cbb7f2355b1d192dbf2935c1f30429bd1f8b6abb8f651
3
+ size 57699778
quantize_config.json CHANGED
@@ -2,87 +2,156 @@
2
  "per_channel": true,
3
  "reduce_range": true,
4
  "per_model_config": {
5
- "encoder_model": {
6
  "op_types": [
7
- "Add",
8
- "Sub",
9
  "Pow",
10
- "Reshape",
11
- "Cast",
12
- "Less",
13
- "Transpose",
14
  "Where",
15
- "ConstantOfShape",
 
 
 
 
 
16
  "Sqrt",
17
  "Mul",
18
  "Gather",
19
- "Div",
20
- "Expand",
21
- "ReduceMean",
22
- "Concat",
23
- "Not",
24
  "Constant",
25
- "Shape",
 
 
 
 
 
26
  "Erf",
27
- "Unsqueeze",
28
- "Softmax",
29
- "Range",
30
- "Slice",
31
- "Equal",
32
- "MatMul"
33
  ],
34
  "weight_type": "QInt8"
35
  },
36
  "decoder_model": {
37
  "op_types": [
38
- "Add",
39
- "Sub",
40
  "Pow",
41
- "Reshape",
42
- "Cast",
43
- "Transpose",
44
  "Where",
45
- "ConstantOfShape",
46
- "Sqrt",
47
  "Sigmoid",
48
- "Relu",
 
 
 
49
  "Mul",
50
  "Gather",
51
- "Div",
 
 
 
 
 
52
  "ReduceSum",
53
- "Expand",
 
 
 
 
 
54
  "ReduceMean",
55
- "Concat",
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
56
  "Clip",
 
 
 
57
  "Abs",
58
- "Constant",
59
- "Dropout",
60
  "Shape",
61
- "Erf",
62
- "Unsqueeze",
 
 
 
 
63
  "Softmax",
64
- "Slice",
65
- "Equal",
66
- "MatMul"
 
 
 
 
 
 
 
 
 
67
  ],
68
  "weight_type": "QInt8"
69
  },
70
- "decoder_postnet_and_vocoder": {
71
  "op_types": [
72
- "Conv",
 
73
  "Add",
74
- "Tanh",
 
75
  "Unsqueeze",
76
- "Squeeze",
 
 
 
 
 
 
 
 
 
 
 
 
77
  "Sub",
78
  "Reshape",
79
- "Transpose",
80
  "Div",
81
- "Constant",
82
- "ConvTranspose",
83
- "LeakyRelu"
 
84
  ],
85
- "weight_type": "QUInt8"
86
  }
87
  }
88
  }
 
2
  "per_channel": true,
3
  "reduce_range": true,
4
  "per_model_config": {
5
+ "decoder_model_merged": {
6
  "op_types": [
7
+ "Expand",
 
8
  "Pow",
9
+ "Add",
10
+ "Unsqueeze",
11
+ "Clip",
 
12
  "Where",
13
+ "Slice",
14
+ "Sigmoid",
15
+ "If",
16
+ "Abs",
17
+ "Shape",
18
+ "Cast",
19
  "Sqrt",
20
  "Mul",
21
  "Gather",
22
+ "Dropout",
23
+ "MatMul",
24
+ "Softmax",
25
+ "Transpose",
 
26
  "Constant",
27
+ "Concat",
28
+ "ReduceSum",
29
+ "Sub",
30
+ "Reshape",
31
+ "Div",
32
+ "ConstantOfShape",
33
  "Erf",
34
+ "Relu",
35
+ "ReduceMean",
36
+ "Equal"
 
 
 
37
  ],
38
  "weight_type": "QInt8"
39
  },
40
  "decoder_model": {
41
  "op_types": [
42
+ "Expand",
 
43
  "Pow",
44
+ "Add",
45
+ "Unsqueeze",
46
+ "Clip",
47
  "Where",
48
+ "Slice",
 
49
  "Sigmoid",
50
+ "Abs",
51
+ "Shape",
52
+ "Cast",
53
+ "Sqrt",
54
  "Mul",
55
  "Gather",
56
+ "Dropout",
57
+ "MatMul",
58
+ "Softmax",
59
+ "Transpose",
60
+ "Constant",
61
+ "Concat",
62
  "ReduceSum",
63
+ "Sub",
64
+ "Reshape",
65
+ "Div",
66
+ "ConstantOfShape",
67
+ "Erf",
68
+ "Relu",
69
  "ReduceMean",
70
+ "Equal"
71
+ ],
72
+ "weight_type": "QInt8"
73
+ },
74
+ "decoder_postnet_and_vocoder": {
75
+ "op_types": [
76
+ "Add",
77
+ "Sub",
78
+ "Conv",
79
+ "Reshape",
80
+ "Tanh",
81
+ "Div",
82
+ "LeakyRelu",
83
+ "Squeeze",
84
+ "Transpose",
85
+ "Constant",
86
+ "Unsqueeze",
87
+ "ConvTranspose"
88
+ ],
89
+ "weight_type": "QUInt8"
90
+ },
91
+ "decoder_with_past_model": {
92
+ "op_types": [
93
+ "Expand",
94
+ "Pow",
95
+ "Add",
96
+ "Unsqueeze",
97
  "Clip",
98
+ "Where",
99
+ "Slice",
100
+ "Sigmoid",
101
  "Abs",
 
 
102
  "Shape",
103
+ "Cast",
104
+ "Sqrt",
105
+ "Mul",
106
+ "Gather",
107
+ "Dropout",
108
+ "MatMul",
109
  "Softmax",
110
+ "Transpose",
111
+ "Constant",
112
+ "Concat",
113
+ "ReduceSum",
114
+ "Sub",
115
+ "Reshape",
116
+ "Div",
117
+ "ConstantOfShape",
118
+ "Erf",
119
+ "Relu",
120
+ "ReduceMean",
121
+ "Equal"
122
  ],
123
  "weight_type": "QInt8"
124
  },
125
+ "encoder_model": {
126
  "op_types": [
127
+ "Expand",
128
+ "Pow",
129
  "Add",
130
+ "Less",
131
+ "Not",
132
  "Unsqueeze",
133
+ "Where",
134
+ "Slice",
135
+ "Shape",
136
+ "Cast",
137
+ "Sqrt",
138
+ "Mul",
139
+ "Gather",
140
+ "MatMul",
141
+ "Softmax",
142
+ "Transpose",
143
+ "Constant",
144
+ "Concat",
145
+ "Range",
146
  "Sub",
147
  "Reshape",
 
148
  "Div",
149
+ "ConstantOfShape",
150
+ "Erf",
151
+ "ReduceMean",
152
+ "Equal"
153
  ],
154
+ "weight_type": "QInt8"
155
  }
156
  }
157
  }
special_tokens_map.json CHANGED
@@ -1,6 +1,13 @@
1
  {
2
  "bos_token": "<s>",
3
  "eos_token": "</s>",
 
 
 
 
 
 
 
4
  "pad_token": "<pad>",
5
  "unk_token": "<unk>"
6
  }
 
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
  }
tokenizer_config.json CHANGED
@@ -1,55 +1,4 @@
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": true,
40
- "single_word": false,
41
- "special": false
42
- },
43
- "80": {
44
- "content": "<ctc_blank>",
45
- "lstrip": true,
46
- "normalized": true,
47
- "rstrip": true,
48
- "single_word": false,
49
- "special": false
50
- }
51
- },
52
- "additional_special_tokens": [],
53
  "bos_token": "<s>",
54
  "clean_up_tokenization_spaces": true,
55
  "eos_token": "</s>",
@@ -59,7 +8,6 @@
59
  "processor_class": "SpeechT5Processor",
60
  "sp_model_kwargs": {},
61
  "tokenizer_class": "SpeechT5Tokenizer",
62
- "tokenizer_file": null,
63
  "trust_remote_code": false,
64
  "unk_token": "<unk>"
65
  }
 
1
  {
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
2
  "bos_token": "<s>",
3
  "clean_up_tokenization_spaces": true,
4
  "eos_token": "</s>",
 
8
  "processor_class": "SpeechT5Processor",
9
  "sp_model_kwargs": {},
10
  "tokenizer_class": "SpeechT5Tokenizer",
 
11
  "trust_remote_code": false,
12
  "unk_token": "<unk>"
13
  }