defkzn commited on
Commit
4e2a62e
·
1 Parent(s): c4b57e9

Upload 3 files

Browse files
Files changed (3) hide show
  1. app.py +6 -3
  2. models/rmedium.onnx +3 -0
  3. models/rmedium.onnx.json +493 -0
app.py CHANGED
@@ -8,20 +8,23 @@ _FILE = Path(__file__)
8
  _DIR = _FILE.parent
9
 
10
  voice = PiperVoice.load(
11
- model_path= _DIR / 'models/ro/ro_RO-mihai-medium.onnx',
12
- config_path= _DIR / 'models/ro/ro_RO-mihai-medium.onnx.json'
13
  )
14
  synthesize_args = {}
15
 
 
16
  def tts(text: str):
 
17
  with tempfile.NamedTemporaryFile(suffix=".wav", delete=False) as fp:
18
  with wave.Wave_write(fp) as wav:
19
  voice.synthesize(text, wav, **synthesize_args)
20
  return fp.name
21
 
 
22
  inputs = [gr.Textbox(label="Input", value="Salut, numele meu este Bogdan.", max_lines=10)]
23
  outputs = gr.Audio(label="Output")
24
 
25
  demo = gr.Interface(fn=tts, inputs=inputs, outputs=outputs)
26
 
27
- demo.launch()
 
8
  _DIR = _FILE.parent
9
 
10
  voice = PiperVoice.load(
11
+ model_path=_DIR / 'models/rmedium.onnx',
12
+ config_path=_DIR / 'models/rmedium.onnx.json'
13
  )
14
  synthesize_args = {}
15
 
16
+
17
  def tts(text: str):
18
+ print(text)
19
  with tempfile.NamedTemporaryFile(suffix=".wav", delete=False) as fp:
20
  with wave.Wave_write(fp) as wav:
21
  voice.synthesize(text, wav, **synthesize_args)
22
  return fp.name
23
 
24
+
25
  inputs = [gr.Textbox(label="Input", value="Salut, numele meu este Bogdan.", max_lines=10)]
26
  outputs = gr.Audio(label="Output")
27
 
28
  demo = gr.Interface(fn=tts, inputs=inputs, outputs=outputs)
29
 
30
+ demo.launch()
models/rmedium.onnx ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:e0608bbbd53c80267c09ece681b09f5199f54e792356684c8073738e5f15d29f
3
+ size 63201294
models/rmedium.onnx.json ADDED
@@ -0,0 +1,493 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "piper_version": "1.0.0",
3
+ "audio": {
4
+ "sample_rate": 22050,
5
+ "quality": "medium"
6
+ },
7
+ "espeak": {
8
+ "voice": "ro"
9
+ },
10
+ "inference": {
11
+ "noise_scale": 0.667,
12
+ "length_scale": 1,
13
+ "noise_w": 0.8
14
+ },
15
+ "phoneme_type": "espeak",
16
+ "phoneme_map": {},
17
+ "phoneme_id_map": {
18
+ "_": [
19
+ 0
20
+ ],
21
+ "^": [
22
+ 1
23
+ ],
24
+ "$": [
25
+ 2
26
+ ],
27
+ " ": [
28
+ 3
29
+ ],
30
+ "!": [
31
+ 4
32
+ ],
33
+ "'": [
34
+ 5
35
+ ],
36
+ "(": [
37
+ 6
38
+ ],
39
+ ")": [
40
+ 7
41
+ ],
42
+ ",": [
43
+ 8
44
+ ],
45
+ "-": [
46
+ 9
47
+ ],
48
+ ".": [
49
+ 10
50
+ ],
51
+ ":": [
52
+ 11
53
+ ],
54
+ ";": [
55
+ 12
56
+ ],
57
+ "?": [
58
+ 13
59
+ ],
60
+ "a": [
61
+ 14
62
+ ],
63
+ "b": [
64
+ 15
65
+ ],
66
+ "c": [
67
+ 16
68
+ ],
69
+ "d": [
70
+ 17
71
+ ],
72
+ "e": [
73
+ 18
74
+ ],
75
+ "f": [
76
+ 19
77
+ ],
78
+ "h": [
79
+ 20
80
+ ],
81
+ "i": [
82
+ 21
83
+ ],
84
+ "j": [
85
+ 22
86
+ ],
87
+ "k": [
88
+ 23
89
+ ],
90
+ "l": [
91
+ 24
92
+ ],
93
+ "m": [
94
+ 25
95
+ ],
96
+ "n": [
97
+ 26
98
+ ],
99
+ "o": [
100
+ 27
101
+ ],
102
+ "p": [
103
+ 28
104
+ ],
105
+ "q": [
106
+ 29
107
+ ],
108
+ "r": [
109
+ 30
110
+ ],
111
+ "s": [
112
+ 31
113
+ ],
114
+ "t": [
115
+ 32
116
+ ],
117
+ "u": [
118
+ 33
119
+ ],
120
+ "v": [
121
+ 34
122
+ ],
123
+ "w": [
124
+ 35
125
+ ],
126
+ "x": [
127
+ 36
128
+ ],
129
+ "y": [
130
+ 37
131
+ ],
132
+ "z": [
133
+ 38
134
+ ],
135
+ "æ": [
136
+ 39
137
+ ],
138
+ "ç": [
139
+ 40
140
+ ],
141
+ "ð": [
142
+ 41
143
+ ],
144
+ "ø": [
145
+ 42
146
+ ],
147
+ "ħ": [
148
+ 43
149
+ ],
150
+ "ŋ": [
151
+ 44
152
+ ],
153
+ "œ": [
154
+ 45
155
+ ],
156
+ "ǀ": [
157
+ 46
158
+ ],
159
+ "ǁ": [
160
+ 47
161
+ ],
162
+ "ǂ": [
163
+ 48
164
+ ],
165
+ "ǃ": [
166
+ 49
167
+ ],
168
+ "ɐ": [
169
+ 50
170
+ ],
171
+ "ɑ": [
172
+ 51
173
+ ],
174
+ "ɒ": [
175
+ 52
176
+ ],
177
+ "ɓ": [
178
+ 53
179
+ ],
180
+ "ɔ": [
181
+ 54
182
+ ],
183
+ "ɕ": [
184
+ 55
185
+ ],
186
+ "ɖ": [
187
+ 56
188
+ ],
189
+ "ɗ": [
190
+ 57
191
+ ],
192
+ "ɘ": [
193
+ 58
194
+ ],
195
+ "ə": [
196
+ 59
197
+ ],
198
+ "ɚ": [
199
+ 60
200
+ ],
201
+ "ɛ": [
202
+ 61
203
+ ],
204
+ "ɜ": [
205
+ 62
206
+ ],
207
+ "ɞ": [
208
+ 63
209
+ ],
210
+ "ɟ": [
211
+ 64
212
+ ],
213
+ "ɠ": [
214
+ 65
215
+ ],
216
+ "ɡ": [
217
+ 66
218
+ ],
219
+ "ɢ": [
220
+ 67
221
+ ],
222
+ "ɣ": [
223
+ 68
224
+ ],
225
+ "ɤ": [
226
+ 69
227
+ ],
228
+ "ɥ": [
229
+ 70
230
+ ],
231
+ "ɦ": [
232
+ 71
233
+ ],
234
+ "ɧ": [
235
+ 72
236
+ ],
237
+ "ɨ": [
238
+ 73
239
+ ],
240
+ "ɪ": [
241
+ 74
242
+ ],
243
+ "ɫ": [
244
+ 75
245
+ ],
246
+ "ɬ": [
247
+ 76
248
+ ],
249
+ "ɭ": [
250
+ 77
251
+ ],
252
+ "ɮ": [
253
+ 78
254
+ ],
255
+ "ɯ": [
256
+ 79
257
+ ],
258
+ "ɰ": [
259
+ 80
260
+ ],
261
+ "ɱ": [
262
+ 81
263
+ ],
264
+ "ɲ": [
265
+ 82
266
+ ],
267
+ "ɳ": [
268
+ 83
269
+ ],
270
+ "ɴ": [
271
+ 84
272
+ ],
273
+ "ɵ": [
274
+ 85
275
+ ],
276
+ "ɶ": [
277
+ 86
278
+ ],
279
+ "ɸ": [
280
+ 87
281
+ ],
282
+ "ɹ": [
283
+ 88
284
+ ],
285
+ "ɺ": [
286
+ 89
287
+ ],
288
+ "ɻ": [
289
+ 90
290
+ ],
291
+ "ɽ": [
292
+ 91
293
+ ],
294
+ "ɾ": [
295
+ 92
296
+ ],
297
+ "ʀ": [
298
+ 93
299
+ ],
300
+ "ʁ": [
301
+ 94
302
+ ],
303
+ "ʂ": [
304
+ 95
305
+ ],
306
+ "ʃ": [
307
+ 96
308
+ ],
309
+ "ʄ": [
310
+ 97
311
+ ],
312
+ "ʈ": [
313
+ 98
314
+ ],
315
+ "ʉ": [
316
+ 99
317
+ ],
318
+ "ʊ": [
319
+ 100
320
+ ],
321
+ "ʋ": [
322
+ 101
323
+ ],
324
+ "ʌ": [
325
+ 102
326
+ ],
327
+ "ʍ": [
328
+ 103
329
+ ],
330
+ "ʎ": [
331
+ 104
332
+ ],
333
+ "ʏ": [
334
+ 105
335
+ ],
336
+ "ʐ": [
337
+ 106
338
+ ],
339
+ "ʑ": [
340
+ 107
341
+ ],
342
+ "ʒ": [
343
+ 108
344
+ ],
345
+ "ʔ": [
346
+ 109
347
+ ],
348
+ "ʕ": [
349
+ 110
350
+ ],
351
+ "ʘ": [
352
+ 111
353
+ ],
354
+ "ʙ": [
355
+ 112
356
+ ],
357
+ "ʛ": [
358
+ 113
359
+ ],
360
+ "ʜ": [
361
+ 114
362
+ ],
363
+ "ʝ": [
364
+ 115
365
+ ],
366
+ "ʟ": [
367
+ 116
368
+ ],
369
+ "ʡ": [
370
+ 117
371
+ ],
372
+ "ʢ": [
373
+ 118
374
+ ],
375
+ "ʲ": [
376
+ 119
377
+ ],
378
+ "ˈ": [
379
+ 120
380
+ ],
381
+ "ˌ": [
382
+ 121
383
+ ],
384
+ "ː": [
385
+ 122
386
+ ],
387
+ "ˑ": [
388
+ 123
389
+ ],
390
+ "˞": [
391
+ 124
392
+ ],
393
+ "β": [
394
+ 125
395
+ ],
396
+ "θ": [
397
+ 126
398
+ ],
399
+ "χ": [
400
+ 127
401
+ ],
402
+ "ᵻ": [
403
+ 128
404
+ ],
405
+ "ⱱ": [
406
+ 129
407
+ ],
408
+ "0": [
409
+ 130
410
+ ],
411
+ "1": [
412
+ 131
413
+ ],
414
+ "2": [
415
+ 132
416
+ ],
417
+ "3": [
418
+ 133
419
+ ],
420
+ "4": [
421
+ 134
422
+ ],
423
+ "5": [
424
+ 135
425
+ ],
426
+ "6": [
427
+ 136
428
+ ],
429
+ "7": [
430
+ 137
431
+ ],
432
+ "8": [
433
+ 138
434
+ ],
435
+ "9": [
436
+ 139
437
+ ],
438
+ "̧": [
439
+ 140
440
+ ],
441
+ "̃": [
442
+ 141
443
+ ],
444
+ "̪": [
445
+ 142
446
+ ],
447
+ "̯": [
448
+ 143
449
+ ],
450
+ "̩": [
451
+ 144
452
+ ],
453
+ "ʰ": [
454
+ 145
455
+ ],
456
+ "ˤ": [
457
+ 146
458
+ ],
459
+ "ε": [
460
+ 147
461
+ ],
462
+ "↓": [
463
+ 148
464
+ ],
465
+ "#": [
466
+ 149
467
+ ],
468
+ "\"": [
469
+ 150
470
+ ],
471
+ "↑": [
472
+ 151
473
+ ],
474
+ "̺": [
475
+ 152
476
+ ],
477
+ "̻": [
478
+ 153
479
+ ]
480
+ },
481
+ "num_symbols": 256,
482
+ "num_speakers": 1,
483
+ "speaker_id_map": {},
484
+ "language": {
485
+ "code": "ro_RO",
486
+ "family": "ro",
487
+ "region": "RO",
488
+ "name_native": "Română",
489
+ "name_english": "Romanian",
490
+ "country_english": "Romania"
491
+ },
492
+ "dataset": "mihai"
493
+ }