-
Notifications
You must be signed in to change notification settings - Fork 0
/
Copy pathindex.html
628 lines (515 loc) · 19.6 KB
/
index.html
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
266
267
268
269
270
271
272
273
274
275
276
277
278
279
280
281
282
283
284
285
286
287
288
289
290
291
292
293
294
295
296
297
298
299
300
301
302
303
304
305
306
307
308
309
310
311
312
313
314
315
316
317
318
319
320
321
322
323
324
325
326
327
328
329
330
331
332
333
334
335
336
337
338
339
340
341
342
343
344
345
346
347
348
349
350
351
352
353
354
355
356
357
358
359
360
361
362
363
364
365
366
367
368
369
370
371
372
373
374
375
376
377
378
379
380
381
382
383
384
385
386
387
388
389
390
391
392
393
394
395
396
397
398
399
400
401
402
403
404
405
406
407
408
409
410
411
412
413
414
415
416
417
418
419
420
421
422
423
424
425
426
427
428
429
430
431
432
433
434
435
436
437
438
439
440
441
442
443
444
445
446
447
448
449
450
451
452
453
454
455
456
457
458
459
460
461
462
463
464
465
466
467
468
469
470
471
472
473
474
475
476
477
478
479
480
481
482
483
484
485
486
487
488
489
490
491
492
493
494
495
496
497
498
499
500
501
502
503
504
505
506
507
508
509
510
511
512
513
514
515
516
517
518
519
520
521
522
523
524
525
526
527
528
529
530
531
532
533
534
535
536
537
538
539
540
541
542
543
544
545
546
547
548
549
550
551
552
553
554
555
556
557
558
559
560
561
562
563
564
565
566
567
568
569
570
571
572
573
574
575
576
577
578
579
580
581
582
583
584
585
586
587
588
589
590
591
592
593
594
595
596
597
598
599
600
601
602
603
604
605
606
607
608
609
610
611
612
613
614
615
616
617
618
619
620
621
622
623
624
625
626
627
<!DOCTYPE html>
<html lang="en-us">
<head>
<meta charset="utf-8">
<meta http-equiv="X-UA-Compatible" content="IE=edge">
<meta name="viewport" content="width=device-width, initial-scale=1, shrink-to-fit=no">
<title> - ICASSP 2020 ESPNET-TTS AUDIO SAMPLES</title>
<meta name="description" content="ICASSP 2020 ESPnet-TTS Audio Samples Abstract This paper introduces a new end-to-end text-to-speech (E2E-TTS) toolkit named ESPnet-TTS, which is an extension of the open-source speech processing toolkit ESPnet. The toolkit supports state-of-the-art E2E-TTS models, including Tacotron 2, Transformer TTS, and FastSpeech, and also provides recipes inspired by the Kaldi automatic speech recognition (ASR) toolkit. The recipes are based on the design unified with the ESPnet ASR recipe, providing high reproducibility.">
<meta name="author" content="">
<link href="https://espnet.github.io/icassp2020-tts/an-old-hope.min.css" rel="stylesheet">
<link href="https://espnet.github.io/icassp2020-tts/style.css" rel="stylesheet">
<link rel="apple-touch-icon" href="https://espnet.github.io/icassp2020-tts/apple-touch-icon.png">
<link rel="icon" href="https://espnet.github.io/icassp2020-tts/favicon.ico">
<meta name="generator" content="Hugo 0.59.0" />
<link rel="alternate" type="application/atom+xml" href="https://espnet.github.io/icassp2020-tts/index.xml" title="ICASSP 2020 ESPNET-TTS AUDIO SAMPLES">
<script>
function setTheme() {
const time = new Date();
const prev = localStorage.getItem('date');
const date = String(time.getMonth() + 1) + '.' + String(time.getDate());
const now = time.getTime();
let sunrise;
let sunset;
function setBodyClass() {
if (now > sunrise && now < sunset) return;
document.body.classList.add('dark');
}
if (date !== prev) {
fetch('https://api.ipgeolocation.io/astronomy?apiKey=5ed37d85103e4defa5df4c5298ed5215')
.then(res => res.json())
.then(data => {
sunrise = data.sunrise.split(':').map(Number);
sunset = data.sunset.split(':').map(Number);
})
.catch(() => {
sunrise = [7, 0];
sunset = [19, 0];
})
.finally(() => {
sunrise = time.setHours(sunrise[0], sunrise[1], 0);
sunset = time.setHours(sunset[0], sunset[1], 0);
setBodyClass();
localStorage.setItem('sunrise', sunrise);
localStorage.setItem('sunset', sunset);
});
localStorage.setItem('date', date);
} else {
sunrise = Number(localStorage.getItem('sunrise'));
sunset = Number(localStorage.getItem('sunset'));
setBodyClass();
}
}
</script>
</head>
<body class="single">
<script>
setTheme();
</script>
<header class="header">
<nav class="nav">
<p class="logo"><a href="https://espnet.github.io/icassp2020-tts/">ICASSP 2020 ESPNET-TTS AUDIO SAMPLES</a></p>
</nav>
</header>
<main class="main">
<article class="post-single">
<header class="post-header">
<h1 class="post-title"></h1>
<div class="post-meta">October 21, 2019</div>
</header>
<div class="post-content">
<h1 id="icassp-2020-espnet-tts-audio-samples">ICASSP 2020 ESPnet-TTS Audio Samples</h1>
<p><a href="https://arxiv.org/abs/1910.10909">https://arxiv.org/abs/1910.10909</a></p>
<h2 id="abstract">Abstract</h2>
<p>This paper introduces a new end-to-end text-to-speech (E2E-TTS) toolkit named ESPnet-TTS, which is an extension of the open-source speech processing toolkit ESPnet. The toolkit supports state-of-the-art E2E-TTS models, including Tacotron 2, Transformer TTS, and FastSpeech, and also provides recipes inspired by the Kaldi automatic speech recognition (ASR) toolkit. The recipes are based on the design unified with the ESPnet ASR recipe, providing high reproducibility. The toolkit also provides pre-trained models and samples of all of the recipes so that users can use it as a baseline. Furthermore, the unified design enables the integration of ASR functions with TTS, e.g., ASR-based objective evaluation and semi-supervised learning with both ASR and TTS models. This paper describes the design of the toolkit and experimental evaluation in comparison with other toolkits. The experimental results show that our best model outperforms other toolkits, resulting in a mean opinion score (MOS) of 4.25 on the LJSpeech dataset. The toolkit is available on <a href="https://github.com/espnet/espnet">GitHub</a>.</p>
<h2 id="audio-samples-english">Audio samples (English)</h2>
<p>These samples are used in the subjective evaluation.<br />
You can find all of the samples in <a href="https://drive.google.com/open?id=1prLlnRmZxN6VRG34KuERaa3u0GySxx7r">our Google Drive</a>.</p>
<h3 id="model-list">Model list</h3>
<ul>
<li><strong>Groundtruth</strong>: Natural speech</li>
<li><strong>Tacotron2.v2</strong>: Tacotron 2 using Forward attention with a transition agent</li>
<li><strong>Tacotron2.v3</strong>: Tacotron 2 with a guided attention loss</li>
<li><strong>Transformer.v1</strong>: Transformer with a guided attention loss</li>
<li><strong>Transformer.v3</strong>: Transformer with a guided attention loss and phoneme inputs</li>
<li><strong>FastSpeech.v2</strong>: FastSpeech trained with Transformer.v1 as a teacher</li>
<li><strong>FastSpeech.v3</strong>: FastSpeech trained with Transformer.v3 as a teacher</li>
<li><strong>FastSpeech.v4</strong>: FastSpeech trained with Transformer.v3 + knowledge distillation (EXPERIMENTAL)</li>
<li><strong>Merlin</strong>: Conventional statistical speech synthesis system based on <a href="https://github.com/CSTR-Edinburgh/merlin">Merlin</a> + <a href="https://github.com/mmorise/World">WORLD</a></li>
<li><strong>Mozilla</strong>: Pre-trained Tacotron 2 and WaveRNN provided by <a href="https://github.com/mozilla/TTS">mozilla/TTS</a></li>
<li><strong>Merlin</strong>: Pre-trained Tacotron 2 and WaveGlow provided by <a href="https://github.com/NVIDIA/tacotron2">NVIDIA/tacotron2</a></li>
</ul>
<h3 id="the-commission-also-recommends"><code>The Commission also recommends</code></h3>
<table>
<thead>
<tr>
<th></th>
<th></th>
<th></th>
</tr>
</thead>
<tbody>
<tr>
<td>Groundtruth</td>
<td>Tacotron2.v2</td>
<td>Tacotron2.v3</td>
</tr>
<tr>
<td><audio controls="" ><source src="wav/ljspeech/raw/LJ050-0030.wav"/></audio></td>
<td><audio controls="" ><source src="wav/ljspeech/tacotron2.v2/LJ050-0030.wav"/></audio></td>
<td><audio controls="" ><source src="wav/ljspeech/tacotron2.v3/LJ050-0030.wav"/></audio></td>
</tr>
<tr>
<td>Transformer.v1</td>
<td>Transformer.v3</td>
<td></td>
</tr>
<tr>
<td><audio controls="" ><source src="wav/ljspeech/transformer.v1/LJ050-0030.wav"/></audio></td>
<td><audio controls="" ><source src="wav/ljspeech/transformer.v3/LJ050-0030.wav"/></audio></td>
<td></td>
</tr>
<tr>
<td>FastSpeech.v2</td>
<td>FastSpeech.v3</td>
<td>FastSpeech.v4</td>
</tr>
<tr>
<td><audio controls="" ><source src="wav/ljspeech/fastspeech.v2/LJ050-0030.wav"/></audio></td>
<td><audio controls="" ><source src="wav/ljspeech/fastspeech.v3/LJ050-0030.wav"/></audio></td>
<td><audio controls="" ><source src="wav/ljspeech/fastspeech.v4/LJ050-0030.wav"/></audio></td>
</tr>
<tr>
<td>Merlin</td>
<td>Mozilla</td>
<td>NVIDIA</td>
</tr>
<tr>
<td><audio controls="" ><source src="wav/ljspeech/merlin/LJ050-0030.wav"/></audio></td>
<td><audio controls="" ><source src="wav/ljspeech/mozilla/LJ050-0030.wav"/></audio></td>
<td><audio controls="" ><source src="wav/ljspeech/nvidia/LJ050-0030.wav"/></audio></td>
</tr>
</tbody>
</table>
<h3 id="as-a-result-of-these-studies-the-planning-document-submitted-by-the-secretary-of-the-treasury-to-the-bureau-of-the-budget-on-august-thirty-one"><code>As a result of these studies, the planning document submitted by the Secretary of the Treasury to the Bureau of the Budget on August thirty-one,</code></h3>
<table>
<thead>
<tr>
<th></th>
<th></th>
<th></th>
</tr>
</thead>
<tbody>
<tr>
<td>Groundtruth</td>
<td>Tacotron2.v2</td>
<td>Tacotron2.v3</td>
</tr>
<tr>
<td><audio controls="" ><source src="wav/ljspeech/raw/LJ050-0050.wav"/></audio></td>
<td><audio controls="" ><source src="wav/ljspeech/tacotron2.v2/LJ050-0050.wav"/></audio></td>
<td><audio controls="" ><source src="wav/ljspeech/tacotron2.v3/LJ050-0050.wav"/></audio></td>
</tr>
<tr>
<td>Transformer.v1</td>
<td>Transformer.v3</td>
<td></td>
</tr>
<tr>
<td><audio controls="" ><source src="wav/ljspeech/transformer.v1/LJ050-0050.wav"/></audio></td>
<td><audio controls="" ><source src="wav/ljspeech/transformer.v3/LJ050-0050.wav"/></audio></td>
<td></td>
</tr>
<tr>
<td>FastSpeech.v2</td>
<td>FastSpeech.v3</td>
<td>FastSpeech.v4</td>
</tr>
<tr>
<td><audio controls="" ><source src="wav/ljspeech/fastspeech.v2/LJ050-0050.wav"/></audio></td>
<td><audio controls="" ><source src="wav/ljspeech/fastspeech.v3/LJ050-0050.wav"/></audio></td>
<td><audio controls="" ><source src="wav/ljspeech/fastspeech.v4/LJ050-0050.wav"/></audio></td>
</tr>
<tr>
<td>Merlin</td>
<td>Mozilla</td>
<td>NVIDIA</td>
</tr>
<tr>
<td><audio controls="" ><source src="wav/ljspeech/merlin/LJ050-0050.wav"/></audio></td>
<td><audio controls="" ><source src="wav/ljspeech/mozilla/LJ050-0050.wav"/></audio></td>
<td><audio controls="" ><source src="wav/ljspeech/nvidia/LJ050-0050.wav"/></audio></td>
</tr>
</tbody>
</table>
<h3 id="the-fbi-now-transmits-information-on-all-defectors-a-category-which-would-of-course-have-included-oswald"><code>The FBI now transmits information on all defectors, a category which would, of course, have included Oswald.</code></h3>
<table>
<thead>
<tr>
<th></th>
<th></th>
<th></th>
</tr>
</thead>
<tbody>
<tr>
<td>Groundtruth</td>
<td>Tacotron2.v2</td>
<td>Tacotron2.v3</td>
</tr>
<tr>
<td><audio controls="" ><source src="wav/ljspeech/raw/LJ050-0070.wav"/></audio></td>
<td><audio controls="" ><source src="wav/ljspeech/tacotron2.v2/LJ050-0070.wav"/></audio></td>
<td><audio controls="" ><source src="wav/ljspeech/tacotron2.v3/LJ050-0070.wav"/></audio></td>
</tr>
<tr>
<td>Transformer.v1</td>
<td>Transformer.v3</td>
<td></td>
</tr>
<tr>
<td><audio controls="" ><source src="wav/ljspeech/transformer.v1/LJ050-0070.wav"/></audio></td>
<td><audio controls="" ><source src="wav/ljspeech/transformer.v3/LJ050-0070.wav"/></audio></td>
<td></td>
</tr>
<tr>
<td>FastSpeech.v2</td>
<td>FastSpeech.v3</td>
<td>FastSpeech.v4</td>
</tr>
<tr>
<td><audio controls="" ><source src="wav/ljspeech/fastspeech.v2/LJ050-0070.wav"/></audio></td>
<td><audio controls="" ><source src="wav/ljspeech/fastspeech.v3/LJ050-0070.wav"/></audio></td>
<td><audio controls="" ><source src="wav/ljspeech/fastspeech.v4/LJ050-0070.wav"/></audio></td>
</tr>
<tr>
<td>Merlin</td>
<td>Mozilla</td>
<td>NVIDIA</td>
</tr>
<tr>
<td><audio controls="" ><source src="wav/ljspeech/merlin/LJ050-0070.wav"/></audio></td>
<td><audio controls="" ><source src="wav/ljspeech/mozilla/LJ050-0070.wav"/></audio></td>
<td><audio controls="" ><source src="wav/ljspeech/nvidia/LJ050-0070.wav"/></audio></td>
</tr>
</tbody>
</table>
<h3 id="they-seem-unduly-restrictive-in-continuing-to-require-some-manifestation-of-animus-against-a-government-official"><code>they seem unduly restrictive in continuing to require some manifestation of animus against a Government official.</code></h3>
<table>
<thead>
<tr>
<th></th>
<th></th>
<th></th>
</tr>
</thead>
<tbody>
<tr>
<td>Groundtruth</td>
<td>Tacotron2.v2</td>
<td>Tacotron2.v3</td>
</tr>
<tr>
<td><audio controls="" ><source src="wav/ljspeech/raw/LJ050-0090.wav"/></audio></td>
<td><audio controls="" ><source src="wav/ljspeech/tacotron2.v2/LJ050-0090.wav"/></audio></td>
<td><audio controls="" ><source src="wav/ljspeech/tacotron2.v3/LJ050-0090.wav"/></audio></td>
</tr>
<tr>
<td>Transformer.v1</td>
<td>Transformer.v3</td>
<td></td>
</tr>
<tr>
<td><audio controls="" ><source src="wav/ljspeech/transformer.v1/LJ050-0090.wav"/></audio></td>
<td><audio controls="" ><source src="wav/ljspeech/transformer.v3/LJ050-0090.wav"/></audio></td>
<td></td>
</tr>
<tr>
<td>FastSpeech.v2</td>
<td>FastSpeech.v3</td>
<td>FastSpeech.v4</td>
</tr>
<tr>
<td><audio controls="" ><source src="wav/ljspeech/fastspeech.v2/LJ050-0090.wav"/></audio></td>
<td><audio controls="" ><source src="wav/ljspeech/fastspeech.v3/LJ050-0090.wav"/></audio></td>
<td><audio controls="" ><source src="wav/ljspeech/fastspeech.v4/LJ050-0090.wav"/></audio></td>
</tr>
<tr>
<td>Merlin</td>
<td>Mozilla</td>
<td>NVIDIA</td>
</tr>
<tr>
<td><audio controls="" ><source src="wav/ljspeech/merlin/LJ050-0090.wav"/></audio></td>
<td><audio controls="" ><source src="wav/ljspeech/mozilla/LJ050-0090.wav"/></audio></td>
<td><audio controls="" ><source src="wav/ljspeech/nvidia/LJ050-0090.wav"/></audio></td>
</tr>
</tbody>
</table>
<h3 id="and-each-agency-given-clear-understanding-of-the-assistance-which-the-secret-service-expects"><code>and each agency given clear understanding of the assistance which the Secret Service expects.</code></h3>
<table>
<thead>
<tr>
<th></th>
<th></th>
<th></th>
</tr>
</thead>
<tbody>
<tr>
<td>Groundtruth</td>
<td>Tacotron2.v2</td>
<td>Tacotron2.v3</td>
</tr>
<tr>
<td><audio controls="" ><source src="wav/ljspeech/raw/LJ050-0110.wav"/></audio></td>
<td><audio controls="" ><source src="wav/ljspeech/tacotron2.v2/LJ050-0110.wav"/></audio></td>
<td><audio controls="" ><source src="wav/ljspeech/tacotron2.v3/LJ050-0110.wav"/></audio></td>
</tr>
<tr>
<td>Transformer.v1</td>
<td>Transformer.v3</td>
<td></td>
</tr>
<tr>
<td><audio controls="" ><source src="wav/ljspeech/transformer.v1/LJ050-0110.wav"/></audio></td>
<td><audio controls="" ><source src="wav/ljspeech/transformer.v3/LJ050-0110.wav"/></audio></td>
<td></td>
</tr>
<tr>
<td>FastSpeech.v2</td>
<td>FastSpeech.v3</td>
<td>FastSpeech.v4</td>
</tr>
<tr>
<td><audio controls="" ><source src="wav/ljspeech/fastspeech.v2/LJ050-0110.wav"/></audio></td>
<td><audio controls="" ><source src="wav/ljspeech/fastspeech.v3/LJ050-0110.wav"/></audio></td>
<td><audio controls="" ><source src="wav/ljspeech/fastspeech.v4/LJ050-0110.wav"/></audio></td>
</tr>
<tr>
<td>Merlin</td>
<td>Mozilla</td>
<td>NVIDIA</td>
</tr>
<tr>
<td><audio controls="" ><source src="wav/ljspeech/merlin/LJ050-0110.wav"/></audio></td>
<td><audio controls="" ><source src="wav/ljspeech/mozilla/LJ050-0110.wav"/></audio></td>
<td><audio controls="" ><source src="wav/ljspeech/nvidia/LJ050-0110.wav"/></audio></td>
</tr>
</tbody>
</table>
<h2 id="audio-samples-mandarin">Audio samples (Mandarin)</h2>
<p>Mandarin samples are created by <a href="https://github.com/espnet/espnet/tree/master/egs/csmsc/tts1"><code>egs/csmsc/tts1</code></a> recipe.<br />
You can find more samples in <a href="https://drive.google.com/open?id=1BmrdQSIatJMZsNMNL64VYARjQXkRrQ1X">our Google Drive</a>.</p>
<h3 id="昨日-这名-伤者-与医生全部被警方依法刑事拘留"><code>昨日，这名“伤者”与医生全部被警方依法刑事拘留。</code></h3>
<table>
<thead>
<tr>
<th></th>
<th></th>
<th></th>
</tr>
</thead>
<tbody>
<tr>
<td>Groundtruth</td>
<td>Transformer.v1</td>
<td>FastSpeech.v3</td>
</tr>
<tr>
<td><audio controls="" ><source src="wav/csmsc/raw/009901.wav"/></audio></td>
<td><audio controls="" ><source src="wav/csmsc/transformer.v1/009901.wav"/></audio></td>
<td><audio controls="" ><source src="wav/csmsc/fastspeech.v3/009901.wav"/></audio></td>
</tr>
</tbody>
</table>
<h3 id="钱伟长想到上海来办学校是经过深思熟虑的"><code>钱伟长想到上海来办学校是经过深思熟虑的。</code></h3>
<table>
<thead>
<tr>
<th></th>
<th></th>
<th></th>
</tr>
</thead>
<tbody>
<tr>
<td>Groundtruth</td>
<td>Transformer.v1</td>
<td>FastSpeech.v3</td>
</tr>
<tr>
<td><audio controls="" ><source src="wav/csmsc/raw/009902.wav"/></audio></td>
<td><audio controls="" ><source src="wav/csmsc/transformer.v1/009902.wav"/></audio></td>
<td><audio controls="" ><source src="wav/csmsc/fastspeech.v3/009902.wav"/></audio></td>
</tr>
</tbody>
</table>
<h3 id="她见我一进门就骂-吃饭时也骂-骂得我抬不起头"><code>她见我一进门就骂，吃饭时也骂，骂得我抬不起头。</code></h3>
<table>
<thead>
<tr>
<th></th>
<th></th>
<th></th>
</tr>
</thead>
<tbody>
<tr>
<td>Groundtruth</td>
<td>Transformer.v1</td>
<td>FastSpeech.v3</td>
</tr>
<tr>
<td><audio controls="" ><source src="wav/csmsc/raw/009903.wav"/></audio></td>
<td><audio controls="" ><source src="wav/csmsc/transformer.v1/009903.wav"/></audio></td>
<td><audio controls="" ><source src="wav/csmsc/fastspeech.v3/009903.wav"/></audio></td>
</tr>
</tbody>
</table>
<h2 id="audio-samples-japanese">Audio samples (Japanese)</h2>
<p>Japanese samples are created by <a href="https://github.com/espnet/espnet/tree/master/egs/jsut/tts1"><code>egs/jsut/tts1</code></a> recipe.<br />
You can find more samples in <a href="https://drive.google.com/open?id=1fEgS4-K4dtgVxwI4Pr7uOA1h4PE-zN7f">our Google Drive</a>.</p>
<h3 id="水をマレーシアから買わなくてはならないのです"><code>水をマレーシアから買わなくてはならないのです。</code></h3>
<table>
<thead>
<tr>
<th></th>
<th></th>
<th></th>
</tr>
</thead>
<tbody>
<tr>
<td>Groundtruth</td>
<td>Tacotron2</td>
<td>Transformer</td>
</tr>
<tr>
<td><audio controls="" ><source src="wav/jsut/raw/BASIC5000_0001.wav"/></audio></td>
<td><audio controls="" ><source src="wav/jsut/tacotron2/BASIC5000_0001.wav"/></audio></td>
<td><audio controls="" ><source src="wav/jsut/transformer/BASIC5000_0001.wav"/></audio></td>
</tr>
</tbody>
</table>
<h3 id="木曜日-停戦会談は-何の進展もないまま終了しました"><code>木曜日、停戦会談は、何の進展もないまま終了しました。</code></h3>
<table>
<thead>
<tr>
<th></th>
<th></th>
<th></th>
</tr>
</thead>
<tbody>
<tr>
<td>Groundtruth</td>
<td>Tacotron2</td>
<td>Transformer</td>
</tr>
<tr>
<td><audio controls="" ><source src="wav/jsut/raw/BASIC5000_0002.wav"/></audio></td>
<td><audio controls="" ><source src="wav/jsut/tacotron2/BASIC5000_0002.wav"/></audio></td>
<td><audio controls="" ><source src="wav/jsut/transformer/BASIC5000_0002.wav"/></audio></td>
</tr>
</tbody>
</table>
<h3 id="上院議員は私がデータをゆがめたと告発した"><code>上院議員は私がデータをゆがめたと告発した。</code></h3>
<table>
<thead>
<tr>
<th></th>
<th></th>
<th></th>
</tr>
</thead>
<tbody>
<tr>
<td>Groundtruth</td>
<td>Tacotron2</td>
<td>Transformer</td>
</tr>
<tr>
<td><audio controls="" ><source src="wav/jsut/raw/BASIC5000_0003.wav"/></audio></td>
<td><audio controls="" ><source src="wav/jsut/tacotron2/BASIC5000_0003.wav"/></audio></td>
<td><audio controls="" ><source src="wav/jsut/transformer/BASIC5000_0003.wav"/></audio></td>
</tr>
</tbody>
</table>
<h2 id="citation">Citation</h2>
<pre><code>@misc{hayashi2019espnettts,
title={ESPnet-TTS: Unified, Reproducible, and Integratable Open Source End-to-End Text-to-Speech Toolkit},
author={Tomoki Hayashi and Ryuichi Yamamoto and Katsuki Inoue and Takenori Yoshimura and Shinji Watanabe and Tomoki Toda and Kazuya Takeda and Yu Zhang and Xu Tan},
year={2019},
eprint={1910.10909},
archivePrefix={arXiv},
primaryClass={cs.CL}
}
</code></pre>
</div>
</article>
</main>
<footer class="footer">
<span>© 2019 <a href="https://espnet.github.io/icassp2020-tts/">ICASSP 2020 ESPNET-TTS AUDIO SAMPLES</a></span>
<span>·</span>
<span>Powered by <a href="https://gohugo.io/" rel="noopener" target="_blank">Hugo️️</a>️</span>
<span>·</span>
<span>Theme️ <a href="https://github.com/nanxiaobei/hugo-paper" rel="noopener" target="_blank">Paper</a></span>
</footer>
<script src="https://espnet.github.io/icassp2020-tts/highlight.min.js"></script>
<script>
hljs.initHighlightingOnLoad();
</script>
</body>
</html>