ONNX
Michael Hansen commited on
Commit
952f41d
1 Parent(s): 2525f22

thorsten_emotional, marylux, dfki, serbski_institut, mihai voices

Browse files
Files changed (30) hide show
  1. _script/voicefest.py +50 -2
  2. de/de_DE/thorsten_emotional/medium/MODEL_CARD +15 -0
  3. de/de_DE/thorsten_emotional/medium/de_DE-thorsten_emotional-medium.onnx +3 -0
  4. de/de_DE/thorsten_emotional/medium/de_DE-thorsten_emotional-medium.onnx.json +502 -0
  5. de/de_DE/thorsten_emotional/medium/samples/speaker_0.mp3 +0 -0
  6. de/de_DE/thorsten_emotional/medium/samples/speaker_1.mp3 +0 -0
  7. de/de_DE/thorsten_emotional/medium/samples/speaker_2.mp3 +0 -0
  8. de/de_DE/thorsten_emotional/medium/samples/speaker_3.mp3 +0 -0
  9. de/de_DE/thorsten_emotional/medium/samples/speaker_4.mp3 +0 -0
  10. de/de_DE/thorsten_emotional/medium/samples/speaker_5.mp3 +0 -0
  11. de/de_DE/thorsten_emotional/medium/samples/speaker_6.mp3 +0 -0
  12. de/de_DE/thorsten_emotional/medium/samples/speaker_7.mp3 +0 -0
  13. lb/lb_LU/marylux/medium/MODEL_CARD +15 -0
  14. lb/lb_LU/marylux/medium/lb_LU-marylux-medium.onnx +3 -0
  15. lb/lb_LU/marylux/medium/lb_LU-marylux-medium.onnx.json +502 -0
  16. lb/lb_LU/marylux/medium/samples/speaker_0.mp3 +0 -0
  17. ro/ro_RO/mihai/medium/MODEL_CARD +15 -0
  18. ro/ro_RO/mihai/medium/ro_RO-mihai-medium.onnx +3 -0
  19. ro/ro_RO/mihai/medium/ro_RO-mihai-medium.onnx.json +493 -0
  20. ro/ro_RO/mihai/medium/samples/speaker_0.mp3 +0 -0
  21. sr/sr_RS/serbski_institut/medium/MODEL_CARD +15 -0
  22. sr/sr_RS/serbski_institut/medium/samples/speaker_0.mp3 +0 -0
  23. sr/sr_RS/serbski_institut/medium/samples/speaker_1.mp3 +0 -0
  24. sr/sr_RS/serbski_institut/medium/sr_RS-serbski_institut-medium.onnx +3 -0
  25. sr/sr_RS/serbski_institut/medium/sr_RS-serbski_institut-medium.onnx.json +505 -0
  26. tr/tr_TR/dfki/medium/MODEL_CARD +15 -0
  27. tr/tr_TR/dfki/medium/samples/speaker_0.mp3 +0 -0
  28. tr/tr_TR/dfki/medium/tr_TR-dfki-medium.onnx +3 -0
  29. tr/tr_TR/dfki/medium/tr_TR-dfki-medium.onnx.json +493 -0
  30. voices.json +162 -0
_script/voicefest.py CHANGED
@@ -32,20 +32,66 @@ _LANGUAGES = {
32
  "it_IT": Language("Italiano", "Italian", "Italy"),
33
  "ka_GE": Language("ქართული ენა", "Georgian", "Georgia"),
34
  "kk_KZ": Language("қазақша", "Kazakh", "Kazakhstan"),
 
35
  "ne_NP": Language("नेपाली", "Nepali", "Nepal"),
36
  "nl_BE": Language("Nederlands", "Dutch", "Belgium"),
37
  "nl_NL": Language("Nederlands", "Dutch", "Netherlands"),
38
  "no_NO": Language("Norsk", "Norwegian", "Norway"),
39
  "pl_PL": Language("Polski", "Polish", "Poland"),
40
  "pt_BR": Language("Português", "Portuguese", "Brazil"),
 
41
  "ru_RU": Language("Русский", "Russian", "Russia"),
 
42
  "sv_SE": Language("Svenska", "Swedish", "Sweden"),
43
  "sw_CD": Language("Kiswahili", "Swahili", "Democratic Republic of the Congo"),
 
44
  "uk_UA": Language("украї́нська мо́ва", "Ukrainian", "Ukraine"),
45
  "vi_VN": Language("Tiếng Việt", "Vietnamese", "Vietnam"),
46
  "zh_CN": Language("简体中文", "Chinese", "China"),
47
  }
48
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
49
 
50
  class VoiceTest(unittest.TestCase):
51
  def test_voices(self):
@@ -109,7 +155,7 @@ class VoiceTest(unittest.TestCase):
109
  )
110
  self.assertEqual(file_dataset, config["dataset"], "Wrong dataset file")
111
  self.assertEqual(
112
- file_quality, config["audio"]["quality"], "Wrong quality file"
113
  )
114
 
115
  # Verify aliases are unique
@@ -128,7 +174,8 @@ class VoiceTest(unittest.TestCase):
128
 
129
  def run_tests() -> None:
130
  runner = unittest.TextTestRunner()
131
- runner.run(unittest.makeSuite(VoiceTest))
 
132
 
133
 
134
  # -----------------------------------------------------------------------------
@@ -236,6 +283,7 @@ def get_file_hash(path, bytes_per_chunk: int = 8192) -> str:
236
  # -----------------------------------------------------------------------------
237
 
238
  if __name__ == "__main__":
 
239
  run_tests()
240
 
241
  print("Writing voices.json")
 
32
  "it_IT": Language("Italiano", "Italian", "Italy"),
33
  "ka_GE": Language("ქართული ენა", "Georgian", "Georgia"),
34
  "kk_KZ": Language("қазақша", "Kazakh", "Kazakhstan"),
35
+ "lb_LU": Language("Lëtzebuergesch", "Luxembourgish", "Luxembourg"),
36
  "ne_NP": Language("नेपाली", "Nepali", "Nepal"),
37
  "nl_BE": Language("Nederlands", "Dutch", "Belgium"),
38
  "nl_NL": Language("Nederlands", "Dutch", "Netherlands"),
39
  "no_NO": Language("Norsk", "Norwegian", "Norway"),
40
  "pl_PL": Language("Polski", "Polish", "Poland"),
41
  "pt_BR": Language("Português", "Portuguese", "Brazil"),
42
+ "ro_RO": Language("Română", "Romanian", "Romania"),
43
  "ru_RU": Language("Русский", "Russian", "Russia"),
44
+ "sr_RS": Language("srpski", "Serbian", "Serbia"),
45
  "sv_SE": Language("Svenska", "Swedish", "Sweden"),
46
  "sw_CD": Language("Kiswahili", "Swahili", "Democratic Republic of the Congo"),
47
+ "tr_TR": Language("Türkçe", "Turkish", "Turkey"),
48
  "uk_UA": Language("украї́нська мо́ва", "Ukrainian", "Ukraine"),
49
  "vi_VN": Language("Tiếng Việt", "Vietnamese", "Vietnam"),
50
  "zh_CN": Language("简体中文", "Chinese", "China"),
51
  }
52
 
53
+ # -----------------------------------------------------------------------------
54
+
55
+
56
+ def add_languages():
57
+ for onnx_path in _REPO_DIR.rglob("*.onnx"):
58
+ config_path = f"{onnx_path}.json"
59
+ with open(config_path, "r", encoding="utf-8") as config_file:
60
+ config = json.load(config_file)
61
+
62
+ lang_code, dataset, quality = onnx_path.stem.split("-")
63
+ is_changed = False
64
+
65
+ if "language" not in config:
66
+ lang_info = _LANGUAGES.get(lang_code)
67
+ assert lang_info is not None, f"Missing name for language: {lang_code}"
68
+
69
+ lang_family, lang_region = lang_code.split("_", maxsplit=1)
70
+ config["language"] = {
71
+ "code": lang_code,
72
+ "family": lang_family,
73
+ "region": lang_region,
74
+ "name_native": lang_info.native,
75
+ "name_english": lang_info.english,
76
+ "country_english": lang_info.country,
77
+ }
78
+ is_changed = True
79
+
80
+ if "dataset" not in config:
81
+ config["dataset"] = dataset
82
+ is_changed = True
83
+
84
+ if "quality" not in config["audio"]:
85
+ config["audio"]["quality"] = quality
86
+ is_changed = True
87
+
88
+ if is_changed:
89
+ with open(config_path, "w", encoding="utf-8") as config_file:
90
+ json.dump(config, config_file, ensure_ascii=False, indent=2)
91
+
92
+
93
+ # -----------------------------------------------------------------------------
94
+
95
 
96
  class VoiceTest(unittest.TestCase):
97
  def test_voices(self):
 
155
  )
156
  self.assertEqual(file_dataset, config["dataset"], "Wrong dataset file")
157
  self.assertEqual(
158
+ file_quality, config["audio"]["quality"], "Wrong quality"
159
  )
160
 
161
  # Verify aliases are unique
 
174
 
175
  def run_tests() -> None:
176
  runner = unittest.TextTestRunner()
177
+ result = runner.run(unittest.makeSuite(VoiceTest))
178
+ assert not result.failures, "Test failures"
179
 
180
 
181
  # -----------------------------------------------------------------------------
 
283
  # -----------------------------------------------------------------------------
284
 
285
  if __name__ == "__main__":
286
+ add_languages()
287
  run_tests()
288
 
289
  print("Writing voices.json")
de/de_DE/thorsten_emotional/medium/MODEL_CARD ADDED
@@ -0,0 +1,15 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ # Model card for thorsten_emotional (medium)
2
+
3
+ * Language: de_DE (German, Germany)
4
+ * Speakers: 8 (emotions)
5
+ * Quality: medium
6
+ * Samplerate: 22,050Hz
7
+
8
+ ## Dataset
9
+
10
+ * URL: https://github.com/thorstenMueller/Thorsten-Voice
11
+ * License: CC0
12
+
13
+ ## Training
14
+
15
+ Finetuned from German thorsten voice (medium quality).
de/de_DE/thorsten_emotional/medium/de_DE-thorsten_emotional-medium.onnx ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:c1764e652266cd6dcebf1b95c61973df5970a5f5272e94b655ff1ddf9a99d1ff
3
+ size 76745905
de/de_DE/thorsten_emotional/medium/de_DE-thorsten_emotional-medium.onnx.json ADDED
@@ -0,0 +1,502 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "piper_version": "1.1.0",
3
+ "audio": {
4
+ "sample_rate": 22050,
5
+ "quality": "medium"
6
+ },
7
+ "espeak": {
8
+ "voice": "de"
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": 8,
483
+ "speaker_id_map": {
484
+ "amused": 0,
485
+ "angry": 1,
486
+ "disgusted": 2,
487
+ "drunk": 3,
488
+ "neutral": 4,
489
+ "sleepy": 5,
490
+ "surprised": 6,
491
+ "whisper": 7
492
+ },
493
+ "language": {
494
+ "code": "de_DE",
495
+ "family": "de",
496
+ "region": "DE",
497
+ "name_native": "Deutsch",
498
+ "name_english": "German",
499
+ "country_english": "Germany"
500
+ },
501
+ "dataset": "thorsten_emotional"
502
+ }
de/de_DE/thorsten_emotional/medium/samples/speaker_0.mp3 ADDED
Binary file (84.8 kB). View file
 
de/de_DE/thorsten_emotional/medium/samples/speaker_1.mp3 ADDED
Binary file (93 kB). View file
 
de/de_DE/thorsten_emotional/medium/samples/speaker_2.mp3 ADDED
Binary file (99.7 kB). View file
 
de/de_DE/thorsten_emotional/medium/samples/speaker_3.mp3 ADDED
Binary file (105 kB). View file
 
de/de_DE/thorsten_emotional/medium/samples/speaker_4.mp3 ADDED
Binary file (86.6 kB). View file
 
de/de_DE/thorsten_emotional/medium/samples/speaker_5.mp3 ADDED
Binary file (123 kB). View file
 
de/de_DE/thorsten_emotional/medium/samples/speaker_6.mp3 ADDED
Binary file (80.4 kB). View file
 
de/de_DE/thorsten_emotional/medium/samples/speaker_7.mp3 ADDED
Binary file (93.6 kB). View file
 
lb/lb_LU/marylux/medium/MODEL_CARD ADDED
@@ -0,0 +1,15 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ # Model card for marylux (medium)
2
+
3
+ * Language: lb_LU (Luxembourgish, Luxembourg)
4
+ * Speakers: 1
5
+ * Quality: medium
6
+ * Samplerate: 22,050Hz
7
+
8
+ ## Dataset
9
+
10
+ * URL: https://github.com/marytts/marylux-data
11
+ * License: http://creativecommons.org/licenses/by-nc-sa/4.0/
12
+
13
+ ## Training
14
+
15
+ Finetuned from U.S. English lessac voice (medium quality).
lb/lb_LU/marylux/medium/lb_LU-marylux-medium.onnx ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:4147ecacdd98932951d0f956555542de358d3ccff708d4996e305c3ce287097a
3
+ size 63201294
lb/lb_LU/marylux/medium/lb_LU-marylux-medium.onnx.json ADDED
@@ -0,0 +1,502 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "piper_version": "1.0.0",
3
+ "audio": {
4
+ "sample_rate": 22050,
5
+ "quality": "medium"
6
+ },
7
+ "espeak": {
8
+ "voice": "lb"
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
+ 3
20
+ ],
21
+ "!": [
22
+ 4
23
+ ],
24
+ "\"": [
25
+ 150
26
+ ],
27
+ "#": [
28
+ 149
29
+ ],
30
+ "$": [
31
+ 2
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
+ "0": [
52
+ 130
53
+ ],
54
+ "1": [
55
+ 131
56
+ ],
57
+ "2": [
58
+ 132
59
+ ],
60
+ "3": [
61
+ 133
62
+ ],
63
+ "4": [
64
+ 134
65
+ ],
66
+ "5": [
67
+ 135
68
+ ],
69
+ "6": [
70
+ 136
71
+ ],
72
+ "7": [
73
+ 137
74
+ ],
75
+ "8": [
76
+ 138
77
+ ],
78
+ "9": [
79
+ 139
80
+ ],
81
+ ":": [
82
+ 11
83
+ ],
84
+ ";": [
85
+ 12
86
+ ],
87
+ "?": [
88
+ 13
89
+ ],
90
+ "X": [
91
+ 156
92
+ ],
93
+ "^": [
94
+ 1
95
+ ],
96
+ "_": [
97
+ 0
98
+ ],
99
+ "a": [
100
+ 14
101
+ ],
102
+ "b": [
103
+ 15
104
+ ],
105
+ "c": [
106
+ 16
107
+ ],
108
+ "d": [
109
+ 17
110
+ ],
111
+ "e": [
112
+ 18
113
+ ],
114
+ "f": [
115
+ 19
116
+ ],
117
+ "g": [
118
+ 154
119
+ ],
120
+ "h": [
121
+ 20
122
+ ],
123
+ "i": [
124
+ 21
125
+ ],
126
+ "j": [
127
+ 22
128
+ ],
129
+ "k": [
130
+ 23
131
+ ],
132
+ "l": [
133
+ 24
134
+ ],
135
+ "m": [
136
+ 25
137
+ ],
138
+ "n": [
139
+ 26
140
+ ],
141
+ "o": [
142
+ 27
143
+ ],
144
+ "p": [
145
+ 28
146
+ ],
147
+ "q": [
148
+ 29
149
+ ],
150
+ "r": [
151
+ 30
152
+ ],
153
+ "s": [
154
+ 31
155
+ ],
156
+ "t": [
157
+ 32
158
+ ],
159
+ "u": [
160
+ 33
161
+ ],
162
+ "v": [
163
+ 34
164
+ ],
165
+ "w": [
166
+ 35
167
+ ],
168
+ "x": [
169
+ 36
170
+ ],
171
+ "y": [
172
+ 37
173
+ ],
174
+ "z": [
175
+ 38
176
+ ],
177
+ "æ": [
178
+ 39
179
+ ],
180
+ "ç": [
181
+ 40
182
+ ],
183
+ "ð": [
184
+ 41
185
+ ],
186
+ "ø": [
187
+ 42
188
+ ],
189
+ "ħ": [
190
+ 43
191
+ ],
192
+ "ŋ": [
193
+ 44
194
+ ],
195
+ "œ": [
196
+ 45
197
+ ],
198
+ "ǀ": [
199
+ 46
200
+ ],
201
+ "ǁ": [
202
+ 47
203
+ ],
204
+ "ǂ": [
205
+ 48
206
+ ],
207
+ "ǃ": [
208
+ 49
209
+ ],
210
+ "ɐ": [
211
+ 50
212
+ ],
213
+ "ɑ": [
214
+ 51
215
+ ],
216
+ "ɒ": [
217
+ 52
218
+ ],
219
+ "ɓ": [
220
+ 53
221
+ ],
222
+ "ɔ": [
223
+ 54
224
+ ],
225
+ "ɕ": [
226
+ 55
227
+ ],
228
+ "ɖ": [
229
+ 56
230
+ ],
231
+ "ɗ": [
232
+ 57
233
+ ],
234
+ "ɘ": [
235
+ 58
236
+ ],
237
+ "ə": [
238
+ 59
239
+ ],
240
+ "ɚ": [
241
+ 60
242
+ ],
243
+ "ɛ": [
244
+ 61
245
+ ],
246
+ "ɜ": [
247
+ 62
248
+ ],
249
+ "ɞ": [
250
+ 63
251
+ ],
252
+ "ɟ": [
253
+ 64
254
+ ],
255
+ "ɠ": [
256
+ 65
257
+ ],
258
+ "ɡ": [
259
+ 66
260
+ ],
261
+ "ɢ": [
262
+ 67
263
+ ],
264
+ "ɣ": [
265
+ 68
266
+ ],
267
+ "ɤ": [
268
+ 69
269
+ ],
270
+ "ɥ": [
271
+ 70
272
+ ],
273
+ "ɦ": [
274
+ 71
275
+ ],
276
+ "ɧ": [
277
+ 72
278
+ ],
279
+ "ɨ": [
280
+ 73
281
+ ],
282
+ "ɪ": [
283
+ 74
284
+ ],
285
+ "ɫ": [
286
+ 75
287
+ ],
288
+ "ɬ": [
289
+ 76
290
+ ],
291
+ "ɭ": [
292
+ 77
293
+ ],
294
+ "ɮ": [
295
+ 78
296
+ ],
297
+ "ɯ": [
298
+ 79
299
+ ],
300
+ "ɰ": [
301
+ 80
302
+ ],
303
+ "ɱ": [
304
+ 81
305
+ ],
306
+ "ɲ": [
307
+ 82
308
+ ],
309
+ "ɳ": [
310
+ 83
311
+ ],
312
+ "ɴ": [
313
+ 84
314
+ ],
315
+ "ɵ": [
316
+ 85
317
+ ],
318
+ "ɶ": [
319
+ 86
320
+ ],
321
+ "ɸ": [
322
+ 87
323
+ ],
324
+ "ɹ": [
325
+ 88
326
+ ],
327
+ "ɺ": [
328
+ 89
329
+ ],
330
+ "ɻ": [
331
+ 90
332
+ ],
333
+ "ɽ": [
334
+ 91
335
+ ],
336
+ "ɾ": [
337
+ 92
338
+ ],
339
+ "ʀ": [
340
+ 93
341
+ ],
342
+ "ʁ": [
343
+ 94
344
+ ],
345
+ "ʂ": [
346
+ 95
347
+ ],
348
+ "ʃ": [
349
+ 96
350
+ ],
351
+ "ʄ": [
352
+ 97
353
+ ],
354
+ "ʈ": [
355
+ 98
356
+ ],
357
+ "ʉ": [
358
+ 99
359
+ ],
360
+ "ʊ": [
361
+ 100
362
+ ],
363
+ "ʋ": [
364
+ 101
365
+ ],
366
+ "ʌ": [
367
+ 102
368
+ ],
369
+ "ʍ": [
370
+ 103
371
+ ],
372
+ "ʎ": [
373
+ 104
374
+ ],
375
+ "ʏ": [
376
+ 105
377
+ ],
378
+ "ʐ": [
379
+ 106
380
+ ],
381
+ "ʑ": [
382
+ 107
383
+ ],
384
+ "ʒ": [
385
+ 108
386
+ ],
387
+ "ʔ": [
388
+ 109
389
+ ],
390
+ "ʕ": [
391
+ 110
392
+ ],
393
+ "ʘ": [
394
+ 111
395
+ ],
396
+ "ʙ": [
397
+ 112
398
+ ],
399
+ "ʛ": [
400
+ 113
401
+ ],
402
+ "ʜ": [
403
+ 114
404
+ ],
405
+ "ʝ": [
406
+ 115
407
+ ],
408
+ "ʟ": [
409
+ 116
410
+ ],
411
+ "ʡ": [
412
+ 117
413
+ ],
414
+ "ʢ": [
415
+ 118
416
+ ],
417
+ "ʦ": [
418
+ 155
419
+ ],
420
+ "ʰ": [
421
+ 145
422
+ ],
423
+ "ʲ": [
424
+ 119
425
+ ],
426
+ "ˈ": [
427
+ 120
428
+ ],
429
+ "ˌ": [
430
+ 121
431
+ ],
432
+ "ː": [
433
+ 122
434
+ ],
435
+ "ˑ": [
436
+ 123
437
+ ],
438
+ "˞": [
439
+ 124
440
+ ],
441
+ "ˤ": [
442
+ 146
443
+ ],
444
+ "̃": [
445
+ 141
446
+ ],
447
+ "̧": [
448
+ 140
449
+ ],
450
+ "̩": [
451
+ 144
452
+ ],
453
+ "̪": [
454
+ 142
455
+ ],
456
+ "̯": [
457
+ 143
458
+ ],
459
+ "̺": [
460
+ 152
461
+ ],
462
+ "̻": [
463
+ 153
464
+ ],
465
+ "β": [
466
+ 125
467
+ ],
468
+ "ε": [
469
+ 147
470
+ ],
471
+ "θ": [
472
+ 126
473
+ ],
474
+ "χ": [
475
+ 127
476
+ ],
477
+ "ᵻ": [
478
+ 128
479
+ ],
480
+ "↑": [
481
+ 151
482
+ ],
483
+ "↓": [
484
+ 148
485
+ ],
486
+ "ⱱ": [
487
+ 129
488
+ ]
489
+ },
490
+ "num_symbols": 256,
491
+ "num_speakers": 1,
492
+ "speaker_id_map": {},
493
+ "language": {
494
+ "code": "lb_LU",
495
+ "family": "lb",
496
+ "region": "LU",
497
+ "name_native": "Lëtzebuergesch",
498
+ "name_english": "Luxembourgish",
499
+ "country_english": "Luxembourg"
500
+ },
501
+ "dataset": "marylux"
502
+ }
lb/lb_LU/marylux/medium/samples/speaker_0.mp3 ADDED
Binary file (17.6 kB). View file
 
ro/ro_RO/mihai/medium/MODEL_CARD ADDED
@@ -0,0 +1,15 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ # Model card for miahi (medium)
2
+
3
+ * Language: ro_RO (Romanian, Romania)
4
+ * Speakers: 1
5
+ * Quality: medium
6
+ * Samplerate: 22,050Hz
7
+
8
+ ## Dataset
9
+
10
+ * URL: https://github.com/NabuCasa/voice-datasets
11
+ * License: CC0
12
+
13
+ ## Training
14
+
15
+ Finetuned from U.S. English lessac voice (medium quality).
ro/ro_RO/mihai/medium/ro_RO-mihai-medium.onnx ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:e0608bbbd53c80267c09ece681b09f5199f54e792356684c8073738e5f15d29f
3
+ size 63201294
ro/ro_RO/mihai/medium/ro_RO-mihai-medium.onnx.json ADDED
@@ -0,0 +1,493 @@