Hugging Face's logo

Datasets:
rhasspy
/
piper-checkpoints

License:
Dataset card Files Community
11
Michael Hansen commited on
Commit
2afd0fc
•
1 Parent(s): 39bcbfd

Add Nepali google

Browse files
Files changed (5) hide show
  1. ne/ne_NP/google/medium/MODEL_CARD +15 -0
  2. ne/ne_NP/google/medium/config.json +501 -0
  3. ne/ne_NP/google/medium/dataset.jsonl +0 -0
  4. ne/ne_NP/google/medium/epoch=2829-step=367900.ckpt +3 -0
  5. ne/ne_NP/google/medium/train.sh +17 -0
ne/ne_NP/google/medium/MODEL_CARD ADDED
@@ -0,0 +1,15 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ # Model card for google (medium)
2
+
3
+ * Language: ne_NP (Nepali, Nepal)
4
+ * Speakers: 18
5
+ * Quality: medium
6
+ * Samplerate: 22,050Hz
7
+
8
+ ## Dataset
9
+
10
+ * URL: http://www.openslr.org/43/
11
+ * License: CC-BY-SA-4.0 International
12
+
13
+ ## Training
14
+
15
+ Finetuned from U.S. English lessac voice (medium quality).
ne/ne_NP/google/medium/config.json ADDED
@@ -0,0 +1,501 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "audio": {
3
+ "sample_rate": 22050
4
+ },
5
+ "espeak": {
6
+ "voice": "ne"
7
+ },
8
+ "inference": {
9
+ "noise_scale": 0.667,
10
+ "length_scale": 1,
11
+ "noise_w": 0.8
12
+ },
13
+ "phoneme_type": "PhonemeType.ESPEAK",
14
+ "phoneme_map": {},
15
+ "phoneme_id_map": {
16
+ "_": [
17
+ 0
18
+ ],
19
+ "^": [
20
+ 1
21
+ ],
22
+ "$": [
23
+ 2
24
+ ],
25
+ " ": [
26
+ 3
27
+ ],
28
+ "!": [
29
+ 4
30
+ ],
31
+ "'": [
32
+ 5
33
+ ],
34
+ "(": [
35
+ 6
36
+ ],
37
+ ")": [
38
+ 7
39
+ ],
40
+ ",": [
41
+ 8
42
+ ],
43
+ "-": [
44
+ 9
45
+ ],
46
+ ".": [
47
+ 10
48
+ ],
49
+ ":": [
50
+ 11
51
+ ],
52
+ ";": [
53
+ 12
54
+ ],
55
+ "?": [
56
+ 13
57
+ ],
58
+ "a": [
59
+ 14
60
+ ],
61
+ "b": [
62
+ 15
63
+ ],
64
+ "c": [
65
+ 16
66
+ ],
67
+ "d": [
68
+ 17
69
+ ],
70
+ "e": [
71
+ 18
72
+ ],
73
+ "f": [
74
+ 19
75
+ ],
76
+ "h": [
77
+ 20
78
+ ],
79
+ "i": [
80
+ 21
81
+ ],
82
+ "j": [
83
+ 22
84
+ ],
85
+ "k": [
86
+ 23
87
+ ],
88
+ "l": [
89
+ 24
90
+ ],
91
+ "m": [
92
+ 25
93
+ ],
94
+ "n": [
95
+ 26
96
+ ],
97
+ "o": [
98
+ 27
99
+ ],
100
+ "p": [
101
+ 28
102
+ ],
103
+ "q": [
104
+ 29
105
+ ],
106
+ "r": [
107
+ 30
108
+ ],
109
+ "s": [
110
+ 31
111
+ ],
112
+ "t": [
113
+ 32
114
+ ],
115
+ "u": [
116
+ 33
117
+ ],
118
+ "v": [
119
+ 34
120
+ ],
121
+ "w": [
122
+ 35
123
+ ],
124
+ "x": [
125
+ 36
126
+ ],
127
+ "y": [
128
+ 37
129
+ ],
130
+ "z": [
131
+ 38
132
+ ],
133
+ "æ": [
134
+ 39
135
+ ],
136
+ "ç": [
137
+ 40
138
+ ],
139
+ "ð": [
140
+ 41
141
+ ],
142
+ "ø": [
143
+ 42
144
+ ],
145
+ "ħ": [
146
+ 43
147
+ ],
148
+ "ŋ": [
149
+ 44
150
+ ],
151
+ "œ": [
152
+ 45
153
+ ],
154
+ "ǀ": [
155
+ 46
156
+ ],
157
+ "ǁ": [
158
+ 47
159
+ ],
160
+ "ǂ": [
161
+ 48
162
+ ],
163
+ "ǃ": [
164
+ 49
165
+ ],
166
+ "ɐ": [
167
+ 50
168
+ ],
169
+ "ɑ": [
170
+ 51
171
+ ],
172
+ "ɒ": [
173
+ 52
174
+ ],
175
+ "ɓ": [
176
+ 53
177
+ ],
178
+ "ɔ": [
179
+ 54
180
+ ],
181
+ "ɕ": [
182
+ 55
183
+ ],
184
+ "ɖ": [
185
+ 56
186
+ ],
187
+ "ɗ": [
188
+ 57
189
+ ],
190
+ "ɘ": [
191
+ 58
192
+ ],
193
+ "ə": [
194
+ 59
195
+ ],
196
+ "ɚ": [
197
+ 60
198
+ ],
199
+ "ɛ": [
200
+ 61
201
+ ],
202
+ "ɜ": [
203
+ 62
204
+ ],
205
+ "ɞ": [
206
+ 63
207
+ ],
208
+ "ɟ": [
209
+ 64
210
+ ],
211
+ "ɠ": [
212
+ 65
213
+ ],
214
+ "ɡ": [
215
+ 66
216
+ ],
217
+ "ɢ": [
218
+ 67
219
+ ],
220
+ "ɣ": [
221
+ 68
222
+ ],
223
+ "ɤ": [
224
+ 69
225
+ ],
226
+ "ɥ": [
227
+ 70
228
+ ],
229
+ "ɦ": [
230
+ 71
231
+ ],
232
+ "ɧ": [
233
+ 72
234
+ ],
235
+ "ɨ": [
236
+ 73
237
+ ],
238
+ "ɪ": [
239
+ 74
240
+ ],
241
+ "ɫ": [
242
+ 75
243
+ ],
244
+ "ɬ": [
245
+ 76
246
+ ],
247
+ "ɭ": [
248
+ 77
249
+ ],
250
+ "ɮ": [
251
+ 78
252
+ ],
253
+ "ɯ": [
254
+ 79
255
+ ],
256
+ "ɰ": [
257
+ 80
258
+ ],
259
+ "ɱ": [
260
+ 81
261
+ ],
262
+ "ɲ": [
263
+ 82
264
+ ],
265
+ "ɳ": [
266
+ 83
267
+ ],
268
+ "ɴ": [
269
+ 84
270
+ ],
271
+ "ɵ": [
272
+ 85
273
+ ],
274
+ "ɶ": [
275
+ 86
276
+ ],
277
+ "ɸ": [
278
+ 87
279
+ ],
280
+ "ɹ": [
281
+ 88
282
+ ],
283
+ "ɺ": [
284
+ 89
285
+ ],
286
+ "ɻ": [
287
+ 90
288
+ ],
289
+ "ɽ": [
290
+ 91
291
+ ],
292
+ "ɾ": [
293
+ 92
294
+ ],
295
+ "ʀ": [
296
+ 93
297
+ ],
298
+ "ʁ": [
299
+ 94
300
+ ],
301
+ "ʂ": [
302
+ 95
303
+ ],
304
+ "ʃ": [
305
+ 96
306
+ ],
307
+ "ʄ": [
308
+ 97
309
+ ],
310
+ "ʈ": [
311
+ 98
312
+ ],
313
+ "ʉ": [
314
+ 99
315
+ ],
316
+ "ʊ": [
317
+ 100
318
+ ],
319
+ "ʋ": [
320
+ 101
321
+ ],
322
+ "ʌ": [
323
+ 102
324
+ ],
325
+ "ʍ": [
326
+ 103
327
+ ],
328
+ "ʎ": [
329
+ 104
330
+ ],
331
+ "ʏ": [
332
+ 105
333
+ ],
334
+ "ʐ": [
335
+ 106
336
+ ],
337
+ "ʑ": [
338
+ 107
339
+ ],
340
+ "ʒ": [
341
+ 108
342
+ ],
343
+ "ʔ": [
344
+ 109
345
+ ],
346
+ "ʕ": [
347
+ 110
348
+ ],
349
+ "ʘ": [
350
+ 111
351
+ ],
352
+ "ʙ": [
353
+ 112
354
+ ],
355
+ "ʛ": [
356
+ 113
357
+ ],
358
+ "ʜ": [
359
+ 114
360
+ ],
361
+ "ʝ": [
362
+ 115
363
+ ],
364
+ "ʟ": [
365
+ 116
366
+ ],
367
+ "ʡ": [
368
+ 117
369
+ ],
370
+ "ʢ": [
371
+ 118
372
+ ],
373
+ "ʲ": [
374
+ 119
375
+ ],
376
+ "ˈ": [
377
+ 120
378
+ ],
379
+ "ˌ": [
380
+ 121
381
+ ],
382
+ "ː": [
383
+ 122
384
+ ],
385
+ "ˑ": [
386
+ 123
387
+ ],
388
+ "˞": [
389
+ 124
390
+ ],
391
+ "β": [
392
+ 125
393
+ ],
394
+ "θ": [
395
+ 126
396
+ ],
397
+ "χ": [
398
+ 127
399
+ ],
400
+ "ᵻ": [
401
+ 128
402
+ ],
403
+ "ⱱ": [
404
+ 129
405
+ ],
406
+ "0": [
407
+ 130
408
+ ],
409
+ "1": [
410
+ 131
411
+ ],
412
+ "2": [
413
+ 132
414
+ ],
415
+ "3": [
416
+ 133
417
+ ],
418
+ "4": [
419
+ 134
420
+ ],
421
+ "5": [
422
+ 135
423
+ ],
424
+ "6": [
425
+ 136
426
+ ],
427
+ "7": [
428
+ 137
429
+ ],
430
+ "8": [
431
+ 138
432
+ ],
433
+ "9": [
434
+ 139
435
+ ],
436
+ "̧": [
437
+ 140
438
+ ],
439
+ "̃": [
440
+ 141
441
+ ],
442
+ "̪": [
443
+ 142
444
+ ],
445
+ "̯": [
446
+ 143
447
+ ],
448
+ "̩": [
449
+ 144
450
+ ],
451
+ "ʰ": [
452
+ 145
453
+ ],
454
+ "ˤ": [
455
+ 146
456
+ ],
457
+ "ε": [
458
+ 147
459
+ ],
460
+ "↓": [
461
+ 148
462
+ ],
463
+ "#": [
464
+ 149
465
+ ],
466
+ "\"": [
467
+ 150
468
+ ],
469
+ "↑": [
470
+ 151
471
+ ],
472
+ "̺": [
473
+ 152
474
+ ],
475
+ "̻": [
476
+ 153
477
+ ]
478
+ },
479
+ "num_symbols": 256,
480
+ "num_speakers": 18,
481
+ "speaker_id_map": {
482
+ "0546": 0,
483
+ "3614": 1,
484
+ "2099": 2,
485
+ "3960": 3,
486
+ "6834": 4,
487
+ "7957": 5,
488
+ "6329": 6,
489
+ "9407": 7,
490
+ "6587": 8,
491
+ "0258": 9,
492
+ "2139": 10,
493
+ "5687": 11,
494
+ "0283": 12,
495
+ "3997": 13,
496
+ "3154": 14,
497
+ "0883": 15,
498
+ "2027": 16,
499
+ "0649": 17
500
+ }
501
+ }
ne/ne_NP/google/medium/dataset.jsonl ADDED
The diff for this file is too large to render. See raw diff
 
ne/ne_NP/google/medium/epoch=2829-step=367900.ckpt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:af9d4946e3a19536d0573d3719afbefaeb57b48f58146717a6bc804ea58d9a5b
3
+ size 924583560
ne/ne_NP/google/medium/train.sh ADDED
@@ -0,0 +1,17 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ #!/usr/bin/env bash
2
+ set -eo pipefail
3
+
4
+ this_dir="$( cd "$( dirname "$0" )" && pwd )"
5
+ repo_dir="$(realpath "${this_dir}/../../../../")"
6
+
7
+ export PYTHONPATH="${repo_dir}/src/python:${PYTHONPATH}"
8
+ python3 -m piper_train \
9
+ --dataset-dir "${this_dir}" \
10
+ --accelerator 'gpu' \
11
+ --devices 1 \
12
+ --batch-size 32 \
13
+ --validation-split 0 \
14
+ --num-test-examples 0 \
15
+ --max_epochs 10000 \
16
+ --checkpoint-epochs 10 \
17
+ --precision 32 "$@"