Hugging Face's logo

thelamapi
/
neuvoice

Text-to-Speech
Transformers
ONNX
PyTorch
text
speech
tts
neuvoice
turkish
turkey
ai voice
text to speech
turkish tts
open source ai
huggingface model
voice cloning
realistic ai voice
speech synthesis
turkish ai
local ai
offline tts
low latency tts
multilingual tts
audio generation
speech model
real time tts
hf model
synthetic voice
Model card Files
xet
 Community
2
Lamapi commited on
Commit
bb23613
·
verified ·
1 Parent(s): c43ad90

Upload 6 files

Browse files
Files changed (6) hide show
  1. onnx/duration_predictor.onnx +3 -0
  2. onnx/text_encoder.onnx +3 -0
  3. onnx/tts.json +311 -0
  4. onnx/unicode_indexer.json +0 -0
  5. onnx/vector_estimator.onnx +3 -0
  6. onnx/vocoder.onnx +3 -0
onnx/duration_predictor.onnx ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:c3eb91414d5ff8a7a239b7fe9e34e7e2bf8a8140d8375ffb14718b1c639325db
3
+ size 3700147
onnx/text_encoder.onnx ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:c7befd5ea8c3119769e8a6c1486c4edc6a3bc8365c67621c881bbb774b9902ff
3
+ size 36416150
onnx/tts.json ADDED
@@ -0,0 +1,311 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "tts_version": "v1.7.3",
3
+ "split": "opensource-multilingual",
4
+ "ttl": {
5
+ "latent_dim": 24,
6
+ "chunk_compress_factor": 6,
7
+ "batch_expander": {
8
+ "n_batch_expand": 6
9
+ },
10
+ "normalizer": {
11
+ "scale": 0.25
12
+ },
13
+ "text_encoder": {
14
+ "n_langs": 0,
15
+ "lang_emb_dim": 0,
16
+ "text_embedder": {
17
+ "char_emb_dim": 256
18
+ },
19
+ "convnext": {
20
+ "idim": 256,
21
+ "ksz": 5,
22
+ "intermediate_dim": 1024,
23
+ "num_layers": 6,
24
+ "dilation_lst": [
25
+ 1,
26
+ 1,
27
+ 2,
28
+ 2,
29
+ 4,
30
+ 4
31
+ ]
32
+ },
33
+ "attn_encoder": {
34
+ "hidden_channels": 256,
35
+ "filter_channels": 1024,
36
+ "n_heads": 4,
37
+ "n_layers": 4,
38
+ "p_dropout": 0.0
39
+ },
40
+ "proj_out": {
41
+ "idim": 256,
42
+ "odim": 256
43
+ }
44
+ },
45
+ "flow_matching": {
46
+ "sig_min": 1e-08
47
+ },
48
+ "style_encoder": {
49
+ "proj_in": {
50
+ "ldim": 24,
51
+ "chunk_compress_factor": 6,
52
+ "odim": 256
53
+ },
54
+ "convnext": {
55
+ "idim": 256,
56
+ "ksz": 5,
57
+ "intermediate_dim": 1024,
58
+ "num_layers": 6,
59
+ "dilation_lst": [
60
+ 1,
61
+ 1,
62
+ 1,
63
+ 1,
64
+ 1,
65
+ 1
66
+ ]
67
+ },
68
+ "style_token_layer": {
69
+ "input_dim": 256,
70
+ "n_style": 50,
71
+ "style_key_dim": 256,
72
+ "style_value_dim": 256,
73
+ "prototype_dim": 256,
74
+ "n_units": 256,
75
+ "n_heads": 2
76
+ }
77
+ },
78
+ "speech_prompted_text_encoder": {
79
+ "text_dim": 256,
80
+ "style_dim": 256,
81
+ "n_units": 256,
82
+ "n_heads": 2
83
+ },
84
+ "uncond_masker": {
85
+ "prob_both_uncond": 0.04,
86
+ "prob_text_uncond": 0.01,
87
+ "std": 0.1,
88
+ "text_dim": 256,
89
+ "n_style": 50,
90
+ "style_key_dim": 256,
91
+ "style_value_dim": 256
92
+ },
93
+ "vector_field": {
94
+ "n_langs": 0,
95
+ "lang_emb_dim": 0,
96
+ "proj_in": {
97
+ "ldim": 24,
98
+ "chunk_compress_factor": 6,
99
+ "odim": 512
100
+ },
101
+ "time_encoder": {
102
+ "time_dim": 64,
103
+ "hdim": 256
104
+ },
105
+ "main_blocks": {
106
+ "n_blocks": 4,
107
+ "time_cond_layer": {
108
+ "idim": 512,
109
+ "time_dim": 64
110
+ },
111
+ "style_cond_layer": {
112
+ "idim": 512,
113
+ "style_dim": 256
114
+ },
115
+ "text_cond_layer": {
116
+ "idim": 512,
117
+ "text_dim": 256,
118
+ "n_heads": 8,
119
+ "n_units": 512,
120
+ "use_residual": true,
121
+ "rotary_base": 10000,
122
+ "rotary_scale": 10
123
+ },
124
+ "convnext_0": {
125
+ "idim": 512,
126
+ "ksz": 5,
127
+ "intermediate_dim": 2048,
128
+ "num_layers": 4,
129
+ "dilation_lst": [
130
+ 1,
131
+ 2,
132
+ 4,
133
+ 8
134
+ ]
135
+ },
136
+ "convnext_1": {
137
+ "idim": 512,
138
+ "ksz": 5,
139
+ "intermediate_dim": 2048,
140
+ "num_layers": 1,
141
+ "dilation_lst": [
142
+ 1
143
+ ]
144
+ },
145
+ "convnext_2": {
146
+ "idim": 512,
147
+ "ksz": 5,
148
+ "intermediate_dim": 2048,
149
+ "num_layers": 1,
150
+ "dilation_lst": [
151
+ 1
152
+ ]
153
+ }
154
+ },
155
+ "last_convnext": {
156
+ "idim": 512,
157
+ "ksz": 5,
158
+ "intermediate_dim": 2048,
159
+ "num_layers": 4,
160
+ "dilation_lst": [
161
+ 1,
162
+ 1,
163
+ 1,
164
+ 1
165
+ ]
166
+ },
167
+ "proj_out": {
168
+ "idim": 512,
169
+ "chunk_compress_factor": 6,
170
+ "ldim": 24
171
+ }
172
+ }
173
+ },
174
+ "ae": {
175
+ "sample_rate": 44100,
176
+ "n_delay": 0,
177
+ "base_chunk_size": 512,
178
+ "chunk_compress_factor": 1,
179
+ "ldim": 24,
180
+ "encoder": {
181
+ "spec_processor": {
182
+ "n_fft": 2048,
183
+ "win_length": 2048,
184
+ "hop_length": 512,
185
+ "n_mels": 228,
186
+ "sample_rate": 44100,
187
+ "eps": 1e-05,
188
+ "norm_mean": 0.0,
189
+ "norm_std": 1.0
190
+ },
191
+ "ksz_init": 7,
192
+ "ksz": 7,
193
+ "num_layers": 10,
194
+ "dilation_lst": [
195
+ 1,
196
+ 1,
197
+ 1,
198
+ 1,
199
+ 1,
200
+ 1,
201
+ 1,
202
+ 1,
203
+ 1,
204
+ 1
205
+ ],
206
+ "intermediate_dim": 2048,
207
+ "idim": 1253,
208
+ "hdim": 512,
209
+ "odim": 24
210
+ },
211
+ "decoder": {
212
+ "ksz_init": 7,
213
+ "ksz": 7,
214
+ "num_layers": 10,
215
+ "dilation_lst": [
216
+ 1,
217
+ 2,
218
+ 4,
219
+ 1,
220
+ 2,
221
+ 4,
222
+ 1,
223
+ 1,
224
+ 1,
225
+ 1
226
+ ],
227
+ "intermediate_dim": 2048,
228
+ "idim": 24,
229
+ "hdim": 512,
230
+ "head": {
231
+ "idim": 512,
232
+ "hdim": 2048,
233
+ "odim": 512,
234
+ "ksz": 3
235
+ }
236
+ }
237
+ },
238
+ "dp": {
239
+ "latent_dim": 24,
240
+ "chunk_compress_factor": 6,
241
+ "normalizer": {
242
+ "scale": 1.0
243
+ },
244
+ "sentence_encoder": {
245
+ "char_emb_dim": 64,
246
+ "text_embedder": {
247
+ "char_emb_dim": 64
248
+ },
249
+ "convnext": {
250
+ "idim": 64,
251
+ "ksz": 5,
252
+ "intermediate_dim": 256,
253
+ "num_layers": 6,
254
+ "dilation_lst": [
255
+ 1,
256
+ 1,
257
+ 1,
258
+ 1,
259
+ 1,
260
+ 1
261
+ ]
262
+ },
263
+ "attn_encoder": {
264
+ "hidden_channels": 64,
265
+ "filter_channels": 256,
266
+ "n_heads": 2,
267
+ "n_layers": 2,
268
+ "p_dropout": 0.0
269
+ },
270
+ "proj_out": {
271
+ "idim": 64,
272
+ "odim": 64
273
+ }
274
+ },
275
+ "style_encoder": {
276
+ "proj_in": {
277
+ "ldim": 24,
278
+ "chunk_compress_factor": 6,
279
+ "odim": 64
280
+ },
281
+ "convnext": {
282
+ "idim": 64,
283
+ "ksz": 5,
284
+ "intermediate_dim": 256,
285
+ "num_layers": 4,
286
+ "dilation_lst": [
287
+ 1,
288
+ 1,
289
+ 1,
290
+ 1
291
+ ]
292
+ },
293
+ "style_token_layer": {
294
+ "input_dim": 64,
295
+ "n_style": 8,
296
+ "style_key_dim": 0,
297
+ "style_value_dim": 16,
298
+ "prototype_dim": 64,
299
+ "n_units": 64,
300
+ "n_heads": 2
301
+ }
302
+ },
303
+ "predictor": {
304
+ "sentence_dim": 64,
305
+ "n_style": 8,
306
+ "style_dim": 16,
307
+ "hdim": 128,
308
+ "n_layer": 2
309
+ }
310
+ }
311
+ }
onnx/unicode_indexer.json ADDED
The diff for this file is too large to render. See raw diff
 
onnx/vector_estimator.onnx ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:883ac868ea0275ef0e991524dc64f16b3c0376efd7c320af6b53f5b780d7c61c
3
+ size 256534781
onnx/vocoder.onnx ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:085de76dd8e8d5836d6ca66826601f615939218f90e519f70ee8a36ed2a4c4ba
3
+ size 101424195