PageSourceSearch

https://ditto-tts.github.io/

html ditto-tts.github.io collected 2026-10-03 09:00:58 UTC 141,495 bytes, 2,084 lines download raw bytes

1<html>
2  <head>
3    <meta charset="utf-8" />
4    <meta name="generator" content="Hugo 0.88.1" />
5    <meta name="viewport" content="width=device-width, initial-scale=1" />
6    <link rel="icon" type="image/x-icon" href="asset/ditto.ico" />
7    <link
8    href="https://fonts.googleapis.com/css?family=Roboto:300,400,700"
9    rel="stylesheet"
10    type="text/css"
11    />
12    <link rel="stylesheet" href="https://cdn.jsdelivr.net/gh/jpswalsh/academicons@1/css/academicons.min.css">
13    <link rel="stylesheet" href="css/custom.css" />
14    <link rel="stylesheet" href="css/normalize.css" />
15
16    <title>DiTTo-TTS</title>
17    <link href="css/bootstrap.min.css" rel="stylesheet" />
18    <link
19      rel="stylesheet"
20      href="https://cdn.jsdelivr.net/npm/[email protected]/font/bootstrap-icons.css"
21    />
22    
22<script
23      src="https://cdn.jsdelivr.net/npm/[email protected]/dist/js/bootstrap.bundle.min.js"
24      integrity="sha384-C6RzsynM9kWDrMNeT87bh95OGNyZPhcTNXj1NW7RuBCsyN/o0jlpcV8Qyq46cDfL"
25      crossorigin="anonymous"
26    ></script>
26
27
28    <link rel="stylesheet" href="css/main.css" />
29    
29<script src="js/main.js" defer></script>
29
30  </head>
31
32  <body data-new-gr-c-s-check-loaded="14.1091.0" data-gr-ext-installed="">
33    <div class="container">
34      <header role="banner"></header>
35      <main role="main">
36        <article itemscope="" itemtype="https://schema.org/BlogPosting">
37          <div class="container pt-5 mt-5 shadow p-5 mb-5 bg-white rounded">
38            <div class="text-center">
39              <h1>DiTTo-TTS:</h1>
40              <h3>
41
42                <span style="font-weight: 760">Di</span>ffusion <span
43                  style="font-weight: 760">T</span>ransformer
44                for Scalable <span
45                style="font-weight: 760">T</span>ext-<span
46                style="font-weight: 760">t</span>o-<span
47                style="font-weight: 760">S</span>peech
48                without Domain-Specific Factors
49              </h3>
50              <p class="lead fw-bold">
51                |<span
52                  style="padding: .375rem .75rem; font-size: 1rem;">ICLR 2025</span>|
53              </p>
54              <div class="affiliation">
55                <span><a target="_blank" href="https://sites.google.com/view/keonlee9420">Keon Lee</a><sup>1</sup></span>,
56                <span><a target="_blank" href="https://github.com/ddwkim">Dong Won Kim</a><sup>1</sup></span>,
57                <span><a target="_blank" href="https://jaywalnut310.github.io/">Jaehyeon Kim</a><sup>2*</sup></span>,
58                <span><a target="_blank" href="https://sites.google.com/view/sjchung">Seungjun Chung</a><sup>1</sup></span>,
59                <span><a target="_blank" href="https://sites.google.com/view/jaewoongcho">Jaewoong Cho</a><sup>1</sup></span>
60                <p style="margin-bottom: 0;"><sup>1</sup><a target="_blank" href="https://www.krafton.ai/en/">KRAFTON</a>, <sup>2</sup><a target="_blank" href="https://www.nvidia.com/en-us/research/">NVIDIA</a></p>
61              </div>
62              <span><small><sup>*</sup>This work was done when Jaehyeon Kim was at KRAFTON.</small></span>
63              <p style="margin-top: 10;" class="links">
64                <a target="_blank" href="https://openreview.net/forum?id=hQvX9MBowC">OpenReview</a>
65                <a target="_blank" href="https://arxiv.org/abs/2406.11427"><span class="icon"><i class="ai ai-arxiv"></i></span> <span>arXiv</span></a>
66              </p>
67            </div>
68            <p>
69              <b>Abstract.</b>
70              Large-scale latent diffusion models (LDMs) excel in content generation
71              across various modalities, but their reliance on phonemes and durations
72              in text-to-speech (TTS) limits scalability and access from other fields.
73              While recent studies show potential in removing these domain-specific
74              factors, performance remains suboptimal. In this work, we introduce DiTTo-TTS,
75              a Diffusion Transformer (DiT)-based TTS model, to investigate whether
76              LDM-based TTS can achieve state-of-the-art performance without domain-specific
77              factors. Through rigorous analysis and empirical exploration, we find that
78              (1) DiT with minimal modifications outperforms U-Net, (2) variable-length
79              modeling with a speech length predictor significantly improves results over
80              fixed-length approaches, and (3) conditions like semantic alignment in speech
81              latent representations are key to further enhancement. By scaling our training
82              data to 82K hours and the model size to 790M parameters, we achieve superior
83              or comparable zero-shot performance to state-of-the-art TTS models in
84              naturalness, intelligibility, and speaker similarity, all without relying
85              on domain-specific factors. Speech samples are available at
86              <a href="https://ditto-tts.github.io">https://ditto-tts.github.io</a>.
87            </p>
88          </div>
89
90          <div class="container pt-5 mt-5 shadow p-5 mb-5 bg-white rounded">
91            <h2 id="model-overview" style="text-align: center">
92              Model Overview
93            </h2>
94
95            <p style="text-align: center">
96              <img src="asset/thisis9.png" height="200" width="800"
97                class="img-fluid" />
98            </p>
99
100            <p>
101              <b>An overview of DiTTo-TTS</b>. (<i>middle</i>) The LDM backbone is trained to
102              denoise a span-masked noisy segment given its contextual counterpart, without utilizing
103              phoneme and phoneme-level duration (see <b>Section 3.1</b>).
104              (<i>left</i>) The inner structure of the DiT block incorporates multi-head cross-attention
105              with global AdaLN (see <b>Section 3.3</b>).
106              (<i>right</i>) The speech length predictor is based on causal transformer blocks (see <b>Section 3.2</b>).
107              Both DiT blocks and the speech length predictor employ cross-attention to condition on text representation.
108              Additionally, DiT blocks utilize AdaLN with the mean-pooled text embedding.
109              '<b>+</b>' denotes addition, and '<b>c</b>' represents concatenation.
110              '<b>text</b>' in sky blue oval represents the embedding from text encoder.
111              The red line indicates a long skip connection between the states before and after the DiT blocks.
112            </p>
113          </div>
114
115          <ul class="nav nav-tabs" id="myTab" role="tablist">
116            <li class="nav-item" role="presentation">
117              <button
118                class="nav-link"
119                data-bs-toggle="tab"
120                data-bs-target="#librispeech-box"
121                type="button"
122                role="tab"
123              >
124                LibriSpeech
125              </button>
126            </li>
127            <li class="nav-item" role="presentation">
128              <button
129                class="nav-link"
130                data-bs-toggle="tab"
131                data-bs-target="#mls-box"
132                type="button"
133                role="tab"
134              >
135                Multilingual
136              </button>
137            </li>
138            <li class="nav-item" role="presentation">
139              <button
140                class="nav-link active"
141                data-bs-toggle="tab"
142                data-bs-target="#celeb-box"
143                type="button"
144                role="tab"
145              >
146                Celebrities
147              </button>
148            </li>
149            <li class="nav-item" role="presentation">
150              <button
151                class="nav-link"
152                data-bs-toggle="tab"
153                data-bs-target="#anime-box"
154                type="button"
155                role="tab"
156              >
157                Anime Characters
158              </button>
159            </li>
160            <li class="nav-item" role="presentation">
161              <button
162                class="nav-link"
163                data-bs-toggle="tab"
164                data-bs-target="#robustness-box"
165                type="button"
166                role="tab"
167              >
168                Robustness
169              </button>
170            </li>
171            <li class="nav-item" role="presentation">
172              <button
173                class="nav-link"
174                data-bs-toggle="tab"
175                data-bs-target="#speech-diversity-box"
176                type="button"
177                role="tab"
178              >
179                Speech Diversity
180              </button>
181            </li>
182            <li class="nav-item" role="presentation">
183              <button
184                class="nav-link"
185                data-bs-toggle="tab"
186                data-bs-target="#speech-rate-contrl-box"
187                type="button"
188                role="tab"
189              >
190                Speech Rate Controllability
191              </button>
192            </li>
193            <li class="nav-item" role="presentation">
194              <button
195                class="nav-link"
196                data-bs-toggle="tab"
197                data-bs-target="#speech-editing-box"
198                type="button"
199                role="tab"
200              >
201                Speech Editing
202              </button>
203            </li>
204            <li class="nav-item" role="presentation">
205              <button
206                class="nav-link"
207                data-bs-toggle="tab"
208                data-bs-target="#compare-box"
209                type="button"
210                role="tab"
211              >
212                Comparison
213              </button>
214            </li>
215          </ul>
216          <div class="tab-content" id="myTabContent">
217            <div
218              id="librispeech-box"
219              role="tabpanel"
220              class="tab-pane fade container pt-5 shadow p-5 mb-5 bg-white rounded"
221            >
222              <h2 id="librispeech" style="text-align: center">
223                Zero-Shot TTS (LibriSpeech)
224              </h2>
225              <p>
226                Through in-context learning, DiTTo-TTS produces speech using a
227                reference audio and its corresponding text. We represent
228                <span style="color: #8fbddf; font-weight: bold"
229                  >the prompt in blue</span
230                >, and
231                <span style="color: #9fde83; font-weight: bold"
232                  >the generated audio in green</span
233                >. The resulting speech closely resembles the reference in
234                voice, background noise, and speaking style. When dividing
235                prompts into 3-second segments, the audio boundaries may not
236                always end in silence and could sometimes cut off words midway.
237                For further details, please see <b>English-only Evaluation</b> part in <b>Section 5.1</b> of our paper.
238                When you click on "Show baselines", you can see the ground truth
239                and baseline audio, etc. YourTTS, Vall-E, and CLaM-TTS audios are brought
240                from YourTTS demo<sup><a href="#footnote1-1">1</a></sup>,
241                 Vall-E demo<sup><a href="#footnote1-2">2</a></sup
242                >, and CLaM-TTS demo<sup><a href="#footnote1-3">3</a></sup
243                  > respectively.
244              </p>
245            </div>
246            <div
247              id="mls-box"
248              role="tabpanel"
249              class="tab-pane fade container pt-5 shadow p-5 mb-5 bg-white rounded"
250            >
251              <h2 id="mls" style="text-align: center">
252                Zero-Shot TTS (Multilingual LibriSpeech)
253              </h2>
254              <p>
255                We represent
256                <span style="color: #8fbddf; font-weight: bold"
257                  >the prompt in blue</span
258                >, and
259                <span style="color: #9fde83; font-weight: bold"
260                  >the generated audio in green</span
261                >. The resulting speech closely resembles the reference in
262                voice, background noise, and speaking style. When dividing
263                prompts into 3-second segments, the audio boundaries may not
264                always end in silence and could sometimes cut off words midway.
265                For further details, please see <b>Multilingual Evaluation</b> part in <b>Section 5.1</b> of our paper.
266                When you click on "Show baselines", you can see the ground truth
267                and baseline audio, etc. CLaM-TTS audios are brought from
268                CLaM-TTS demo<sup><a href="#footnote2-1">1</a></sup>.
269              </p>
270            </div>
271            <div
272              id="celeb-box"
273              role="tabpanel"
274              class="tab-pane fade show active container pt-5 shadow p-5 mb-5 bg-white rounded"
275            >
276              <h2 id="celeb" style="text-align: center">Celebrities</h2>
277              <p>
278                DiTTo-TTS is capable of replicating the voices and speech styles of well-known personalities.
279                We represent
280                <span style="color: #8fbddf; font-weight: bold"
281                  >the prompt in blue</span
282                >, and
283                <span style="color: #9fde83; font-weight: bold"
284                  >the generated audio in green</span
285                >. The resulting speech closely resembles the reference in
286                voice, background noise, and speaking style. When you click on
287                "Show baselines", you can see the ground truth and baseline
288                audio, etc. Texts corresponding to the generated audio and CLaM-TTS audios are brought from
289                CLaM-TTS demo<sup><a href="#footnote3-1">1</a></sup>.
290              </p>
291            </div>
292            <div
293              id="anime-box"
294              role="tabpanel"
295              class="tab-pane fade container pt-5 shadow p-5 mb-5 bg-white rounded"
296            >
297              <h2 id="anime" style="text-align: center">Anime Characters</h2>
298              <p>
299                DiTTo-TTS is capable of replicating the voices and speech styles of well-known anime characters.
300                We represent
301                <span style="color: #8fbddf; font-weight: bold"
302                  >the prompt in blue</span
303                >, and
304                <span style="color: #9fde83; font-weight: bold"
305                  >the generated audio in green</span
306                >. The resulting speech closely resembles the reference in
307                voice, background noise, and speaking style. When you click on
308                "Show baselines", you can see the ground truth and baseline
309                audio, etc. Texts corresponding to the generated audio and Mega-TTS audios are brought from
310                Mega-TTS demo<sup><a href="#footnote4-1">1</a></sup>.
311              </p>
312            </div>
313            <div
314              id="robustness-box"
315              role="tabpanel"
316              class="tab-pane fade container pt-5 shadow p-5 mb-5 bg-white rounded"
317            >
318              <h2 id="robustness" style="text-align: center">Robustness</h2>
319              <p>
320                DiTTo-TTS can generate robust speech, as demonstrated by its low WER.
321                Additionally, our model is capable of consistently producing a
322                '<b>whispering</b>' effect (please listen to the first sample).
323                Baseline samples are taken from Mega-TTS demo<sup
324                  ><a href="#footnote5-1">1</a></sup
325                > and CLaM-TTS demo<sup
326                ><a href="#footnote5-2">2</a></sup
327              >.
328              </p>
329              <div class="card card-body ditto-card">
330                <div class="ditto-sample-box">
331                  <div class="table-responsive pt-3">
332                    <table class="table table-hover pt-2">
333                      <thead>
334                        <tr>
335                          <th
336                            style="
337                              width: 500px;
338                              max-width: 100%;
339                              text-align: center;
340                            "
341                          >
342                            Text
343                          </th>
344                          <th style="text-align: center">Mega-TTS</th>
345                          <th style="text-align: center">CLaM-TTS</th>
346                          <th style="text-align: center">DiTTo-TTS</th>
347                        </tr>
348                      </thead>
349                      <tbody>
350                        <tr>
351                          <td
352                            data-label="Text"
353                            rowspan="1"
354                            style="
355                              text-align: left;
356                              vertical-align: middle;
357                              max-width: 100%;
358                            "
359                          >
360                            Thursday, via a joint press release and Microsoft
361                            speech Blog, we will announce Microsoft’s continued
362                            partnership with Shell leveraging cloud, speech, and
363                            collaboration technology to drive industry
364                            innovation and transformation.
365                          </td>
366                          <td data-label="Mega-TTS" style="text-align: center">
367                            <audio
368                              controls="controls"
369                              preload="none"
370                              style="width: 140px"
371                            >
372                              <source src="audios/robust/1_mega.wav" autoplay />
373                              Your browser does not support the audio element.
374                            </audio>
375                          </td>
376                          <td data-label="CLaM-TTS" style="text-align: center">
377                            <audio
378                              controls="controls"
379                              preload="none"
380                              style="width: 140px"
381                            >
382                              <source src="audios/robust/1_clam.wav" autoplay />
383                              Your browser does not support the audio element.
384                            </audio>
385                          </td>
386                          <td data-label="DiTTo-TTS" style="text-align: center">
387                            <audio
388                              controls="controls"
389                              preload="none"
390                              style="width: 140px"
391                            >
392                              <source src="audios/robust/1_ours.wav" autoplay />
393                              Your browser does not support the audio element.
394                            </audio>
395                          </td>
396                        </tr>
397                        <tr>
398                          <td
399                            data-label="Text"
400                            rowspan="1"
401                            style="
402                              text-align: left;
403                              vertical-align: middle;
404                              max-width: 100%;
405                            "
406                          >
407                            The great Greek grape growers grow great Greek
408                            grapes one one one.
409                          </td>
410                          <td data-label="Mega-TTS" style="text-align: center">
411                            <audio
412                              controls="controls"
413                              preload="none"
414                              style="width: 140px"
415                            >
416                              <source src="audios/robust/0_mega.wav" autoplay />
417                              Your browser does not support the audio element.
418                            </audio>
419                          </td>
420                          <td data-label="CLaM-TTS" style="text-align: center">
421                            <audio
422                              controls="controls"
423                              preload="none"
424                              style="width: 140px"
425                            >
426                              <source src="audios/robust/0_clam.wav" autoplay />
427                              Your browser does not support the audio element.
428                            </audio>
429                          </td>
430                          <td data-label="DiTTo-TTS" style="text-align: center">
431                            <audio
432                              controls="controls"
433                              preload="none"
434                              style="width: 140px"
435                            >
436                              <source src="audios/robust/0_ours.wav" autoplay />
437                              Your browser does not support the audio element.
438                            </audio>
439                          </td>
440                        </tr>
441                      </tbody>
442                    </table>
443                  </div>
444                </div>
445              </div>
446              <br>
447              <br>
448              <h2 id="robustness" style="text-align: center">Comparision Results on RAVDESS Benchmark</h2>
449              <p>
450                The RAVDESS dataset includes two specific textual prompts: “Dogs are sitting by the door” used for the prompt text, and “Kids are talking by the door” utilized for the synthesis text.
451                Baseline samples are taken from the NaturalSpeech 3 demo<sup
452                  ><a href="#footnote5-3">3</a></sup
453                >. For more details, refer to <b>Section 5.1</b> of our paper.
454              </p>
455              <div class="card card-body ditto-card">
456                <div class="ditto-sample-box">
457                  <div class="table-responsive pt-3">
458                    <table class="table table-hover pt-2">
459                      <thead>
460                      <tr>
461                      <th style="text-align: center">Prompt Emotion</th>
462                      <th style="text-align: center">Prompt</th>
463                      <th style="text-align: center">Ground Truth</th>
464                      <th style="text-align: center">DiTTo-TTS</th>
465                      <th style="text-align: center">NaturalSpeech 3</th>
466                      <th style="text-align: center">NaturalSpeech 2</th>
467                      <th style="text-align: center">Voicebox (R)</th>
468                      <th style="text-align: center">VALL-E (R)</th>
469                      <th style="text-align: center">Mega-TTS 2</th>
470                      <th style="text-align: center">StyleTTS 2</th>
471                      <th style="text-align: center">HierSpeech++</th>
472                      </tr>
473                    </thead>
474                    <tbody>
475                      <tr>
476                        <td style="text-align: center;vertical-align:middle;width: 200px;min-width: 200px">neutral</td>
477                        <td style="text-align: center"><audio controls="controls" style="width: 160px;"><source src="audios/ravdess/2_prompt_1.wav">Your browser does not support the audio element.</audio></td>
478                        <td style="text-align: center"><audio controls="controls" style="width: 160px;"><source src="audios/ravdess/2_gt_1.wav">Your browser does not support the audio element.</audio></td>
479                        <td style="text-align: center"><audio controls="controls" style="width: 160px;"><source src="audios/ravdess/gt/rvd_bm_0_neutral_ours.wav">Your browser does not support the audio element.</audio></td>
480                        <td style="text-align: center"><audio controls="controls" style="width: 160px;"><source src="audios/ravdess/ns3/rvd_bm_0_neutral.wav">Your browser does not support the audio element.</audio></td>
481                        <td style="text-align: center"><audio controls="controls" style="width: 160px;"><source src="audios/ravdess/ns2/rvd_bm_0_neutral.wav">Your browser does not support the audio element.</audio></td>
482                        <td style="text-align: center"><audio controls="controls" style="width: 160px;"><source src="audios/ravdess/voiceboxr/rvd_bm_0_neutral.wav">Your browser does not support the audio element.</audio></td>
483                        <td style="text-align: center"><audio controls="controls" style="width: 160px;"><source src="audios/ravdess/valler/rvd_bm_0_neutral.wav">Your browser does not support the audio element.</audio></td>
484                        <td style="text-align: center"><audio controls="controls" style="width: 160px;"><source src="audios/ravdess/megatts2/rvd_bm_0_neutral.wav">Your browser does not support the audio element.</audio></td>
485                        <td style="text-align: center"><audio controls="controls" style="width: 160px;"><source src="audios/ravdess/styletts2/rvd_bm_0_neutral.wav">Your browser does not support the audio element.</audio></td>
486                        <td style="text-align: center"><audio controls="controls" style="width: 160px;"><source src="audios/ravdess/hierspeechpp/rvd_bm_0_neutral.wav">Your browser does not support the audio element.</audio></td>
487                      </tr>
488            
489                      <tr>
490                        <td style="text-align: center;vertical-align:middle;width: 200px;min-width: 200px">happy</td>
491                        <td style="text-align: center"><audio controls="controls" style="width: 160px;"><source src="audios/ravdess/2_prompt_2.wav">Your browser does not support the audio element.</audio></td>
492                        <td style="text-align: center"><audio controls="controls" style="width: 160px;"><source src="audios/ravdess/2_gt_2.wav">Your browser does not support the audio element.</audio></td>
493                        <td style="text-align: center"><audio controls="controls" style="width: 160px;"><source src="audios/ravdess/gt/rvd_bm_1_happy_ours.wav">Your browser does not support the audio element.</audio></td>
494                        <td style="text-align: center"><audio controls="controls" style="width: 160px;"><source src="audios/ravdess/ns3/rvd_bm_1_happy.wav">Your browser does not support the audio element.</audio></td>
495                        <td style="text-align: center"><audio controls="controls" style="width: 160px;"><source src="audios/ravdess/ns2/rvd_bm_1_happy.wav">Your browser does not support the audio element.</audio></td>
496                        <td style="text-align: center"><audio controls="controls" style="width: 160px;"><source src="audios/ravdess/voiceboxr/rvd_bm_1_happy.wav">Your browser does not support the audio element.</audio></td>
497                        <td style="text-align: center"><audio controls="controls" style="width: 160px;"><source src="audios/ravdess/valler/rvd_bm_1_happy.wav">Your browser does not support the audio element.</audio></td>
498                        <td style="text-align: center"><audio controls="controls" style="width: 160px;"><source src="audios/ravdess/megatts2/rvd_bm_1_happy.wav">Your browser does not support the audio element.</audio></td>
499                        <td style="text-align: center"><audio controls="controls" style="width: 160px;"><source src="audios/ravdess/styletts2/rvd_bm_1_happy.wav">Your browser does not support the audio element.</audio></td>
500                        <td style="text-align: center"><audio controls="controls" style="width: 160px;"><source src="audios/ravdess/hierspeechpp/rvd_bm_1_happy.wav">Your browser does not support the audio element.</audio></td>
501                      </tr>
502            
503                      <tr>
504                        <td style="text-align: center;vertical-align:middle;width: 200px;min-width: 200px">calm</td>
505                        <td style="text-align: center"><audio controls="controls" style="width: 160px;"><source src="audios/ravdess/2_prompt_3.wav">Your browser does not support the audio element.</audio></td>
506                        <td style="text-align: center"><audio controls="controls" style="width: 160px;"><source src="audios/ravdess/2_gt_3.wav">Your browser does not support the audio element.</audio></td>
507                        <td style="text-align: center"><audio controls="controls" style="width: 160px;"><source src="audios/ravdess/gt/rvd_bm_2_calm_ours.wav">Your browser does not support the audio element.</audio></td>
508                        <td style="text-align: center"><audio controls="controls" style="width: 160px;"><source src="audios/ravdess/ns3/rvd_bm_2_calm.wav">Your browser does not support the audio element.</audio></td>
509                        <td style="text-align: center"><audio controls="controls" style="width: 160px;"><source src="audios/ravdess/ns2/rvd_bm_2_calm.wav">Your browser does not support the audio element.</audio></td>
510                        <td style="text-align: center"><audio controls="controls" style="width: 160px;"><source src="audios/ravdess/voiceboxr/rvd_bm_2_calm.wav">Your browser does not support the audio element.</audio></td>
511                        <td style="text-align: center"><audio controls="controls" style="width: 160px;"><source src="audios/ravdess/valler/rvd_bm_2_calm.wav">Your browser does not support the audio element.</audio></td>
512                        <td style="text-align: center"><audio controls="controls" style="width: 160px;"><source src="audios/ravdess/megatts2/rvd_bm_2_calm.wav">Your browser does not support the audio element.</audio></td>
513                        <td style="text-align: center"><audio controls="controls" style="width: 160px;"><source src="audios/ravdess/styletts2/rvd_bm_2_calm.wav">Your browser does not support the audio element.</audio></td>
514                        <td style="text-align: center"><audio controls="controls" style="width: 160px;"><source src="audios/ravdess/hierspeechpp/rvd_bm_2_calm.wav">Your browser does not support the audio element.</audio></td>
515                      </tr>
516            
517            
518                      <tr>
519                        <td style="text-align: center;vertical-align:middle;width: 200px;min-width: 200px">sad</td>
520                        <td style="text-align: center"><audio controls="controls" style="width: 160px;"><source src="audios/ravdess/2_prompt_4.wav">Your browser does not support the audio element.</audio></td>
521                        <td style="text-align: center"><audio controls="controls" style="width: 160px;"><source src="audios/ravdess/2_gt_4.wav">Your browser does not support the audio element.</audio></td>
522                        <td style="text-align: center">
522<audio controls="controls" style="width: 160px;"><source src="audios/ravdess/gt/rvd_bm_3_sad_ours.wav">Your browser does not support the audio element.</audio></td>
523                        <td style="text-align: center"><audio controls="controls" style="width: 160px;"><source src="audios/ravdess/ns3/rvd_bm_3_sad.wav">Your browser does not support the audio element.</audio></td>
524                        <td style="text-align: center"><audio controls="controls" style="width: 160px;"><source src="audios/ravdess/ns2/rvd_bm_3_sad.wav">Your browser does not support the audio element.</audio></td>
525                        <td style="text-align: center"><audio controls="controls" style="width: 160px;"><source src="audios/ravdess/voiceboxr/rvd_bm_3_sad.wav">Your browser does not support the audio element.</audio></td>
526                        <td style="text-align: center"><audio controls="controls" style="width: 160px;"><source src="audios/ravdess/valler/rvd_bm_3_sad.wav">Your browser does not support the audio element.</audio></td>
527                        <td style="text-align: center"><audio controls="controls" style="width: 160px;"><source src="audios/ravdess/megatts2/rvd_bm_3_sad.wav">Your browser does not support the audio element.</audio></td>
528                        <td style="text-align: center"><audio controls="controls" style="width: 160px;"><source src="audios/ravdess/styletts2/rvd_bm_3_sad.wav">Your browser does not support the audio element.</audio></td>
529                        <td style="text-align: center"><audio controls="controls" style="width: 160px;"><source src="audios/ravdess/hierspeechpp/rvd_bm_3_sad.wav">Your browser does not support the audio element.</audio></td>
530                      </tr>
531                    
532                      <tr>
533                        <td style="text-align: center;vertical-align:middle;width: 200px;min-width: 200px">angry</td>
534                        <td style="text-align: center"><audio controls="controls" style="width: 160px;"><source src="audios/ravdess/2_prompt_5.wav">Your browser does not support the audio element.</audio></td>
535                        <td style="text-align: center"><audio controls="controls" style="width: 160px;"><source src="audios/ravdess/2_gt_5.wav">Your browser does not support the audio element.</audio></td>
536                        <td style="text-align: center"><audio controls="controls" style="width: 160px;"><source src="audios/ravdess/gt/rvd_bm_4_angry_ours.wav">Your browser does not support the audio element.</audio></td>
537                        <td style="text-align: center"><audio controls="controls" style="width: 160px;"><source src="audios/ravdess/ns3/rvd_bm_4_angry.wav">Your browser does not support the audio element.</audio></td>
538                        <td style="text-align: center"><audio controls="controls" style="width: 160px;"><source src="audios/ravdess/ns2/rvd_bm_4_angry.wav">Your browser does not support the audio element.</audio></td>
539                        <td style="text-align: center"><audio controls="controls" style="width: 160px;"><source src="audios/ravdess/voiceboxr/rvd_bm_4_angry.wav">Your browser does not support the audio element.</audio></td>
540                        <td style="text-align: center"><audio controls="controls" style="width: 160px;"><source src="audios/ravdess/valler/rvd_bm_4_angry.wav">Your browser does not support the audio element.</audio></td>
541                        <td style="text-align: center"><audio controls="controls" style="width: 160px;"><source src="audios/ravdess/megatts2/rvd_bm_4_angry.wav">Your browser does not support the audio element.</audio></td>
542                        <td style="text-align: center"><audio controls="controls" style="width: 160px;"><source src="audios/ravdess/styletts2/rvd_bm_4_angry.wav">Your browser does not support the audio element.</audio></td>
543                        <td style="text-align: center"><audio controls="controls" style="width: 160px;"><source src="audios/ravdess/hierspeechpp/rvd_bm_4_angry.wav">Your browser does not support the audio element.</audio></td>
544                      </tr>
545            
546                      <tr>
547                        <td style="text-align: center;vertical-align:middle;width: 200px;min-width: 200px">fearful</td>
548                        <td style="text-align: center"><audio controls="controls" style="width: 160px;"><source src="audios/ravdess/2_prompt_6.wav">Your browser does not support the audio element.</audio></td>
549                        <td style="text-align: center"><audio controls="controls" style="width: 160px;"><source src="audios/ravdess/2_gt_6.wav">Your browser does not support the audio element.</audio></td>
550                        <td style="text-align: center"><audio controls="controls" style="width: 160px;"><source src="audios/ravdess/gt/rvd_bm_5_fearful_ours.wav">Your browser does not support the audio element.</audio></td>
551                        <td style="text-align: center"><audio controls="controls" style="width: 160px;"><source src="audios/ravdess/ns3/rvd_bm_5_fearful.wav">Your browser does not support the audio element.</audio></td>
552                        <td style="text-align: center"><audio controls="controls" style="width: 160px;"><source src="audios/ravdess/ns2/rvd_bm_5_fearful.wav">Your browser does not support the audio element.</audio></td>
553                        <td style="text-align: center"><audio controls="controls" style="width: 160px;"><source src="audios/ravdess/voiceboxr/rvd_bm_5_fearful.wav">Your browser does not support the audio element.</audio></td>
554                        <td style="text-align: center"><audio controls="controls" style="width: 160px;"><source src="audios/ravdess/valler/rvd_bm_5_fearful.wav">Your browser does not support the audio element.</audio></td>
555                        <td style="text-align: center"><audio controls="controls" style="width: 160px;"><source src="audios/ravdess/megatts2/rvd_bm_5_fearful.wav">Your browser does not support the audio element.</audio></td>
556                        <td style="text-align: center"><audio controls="controls" style="width: 160px;"><source src="audios/ravdess/styletts2/rvd_bm_5_fearful.wav">Your browser does not support the audio element.</audio></td>
557                        <td style="text-align: center"><audio controls="controls" style="width: 160px;"><source src="audios/ravdess/hierspeechpp/rvd_bm_5_fearful.wav">Your browser does not support the audio element.</audio></td>
558                      </tr>
559                      
560                      <tr>
561                        <td style="text-align: center;vertical-align:middle;width: 200px;min-width: 200px">disgust</td>
562                        <td style="text-align: center"><audio controls="controls" style="width: 160px;"><source src="audios/ravdess/2_prompt_7.wav">Your browser does not support the audio element.</audio></td>
563                        <td style="text-align: center"><audio controls="controls" style="width: 160px;"><source src="audios/ravdess/2_gt_7.wav">Your browser does not support the audio element.</audio></td>
564                        <td style="text-align: center"><audio controls="controls" style="width: 160px;"><source src="audios/ravdess/gt/rvd_bm_6_disgust_ours.wav">Your browser does not support the audio element.</audio></td>
565                        <td style="text-align: center"><audio controls="controls" style="width: 160px;"><source src="audios/ravdess/ns3/rvd_bm_6_disgust.wav">Your browser does not support the audio element.</audio></td>
566                        <td style="text-align: center"><audio controls="controls" style="width: 160px;"><source src="audios/ravdess/ns2/rvd_bm_6_disgust.wav">Your browser does not support the audio element.</audio></td>
567                        <td style="text-align: center"><audio controls="controls" style="width: 160px;"><source src="audios/ravdess/voiceboxr/rvd_bm_6_disgust.wav">Your browser does not support the audio element.</audio></td>
568                        <td style="text-align: center"><audio controls="controls" style="width: 160px;"><source src="audios/ravdess/valler/rvd_bm_6_disgust.wav">Your browser does not support the audio element.</audio></td>
569                        <td style="text-align: center"><audio controls="controls" style="width: 160px;"><source src="audios/ravdess/megatts2/rvd_bm_6_disgust.wav">Your browser does not support the audio element.</audio></td>
570                        <td style="text-align: center"><audio controls="controls" style="width: 160px;"><source src="audios/ravdess/styletts2/rvd_bm_6_disgust.wav">Your browser does not support the audio element.</audio></td>
571                        <td style="text-align: center"><audio controls="controls" style="width: 160px;"><source src="audios/ravdess/hierspeechpp/rvd_bm_6_disgust.wav">Your browser does not support the audio element.</audio></td>
572                      </tr>
573            
574                      <tr>
575                        <td style="text-align: center;vertical-align:middle;width: 200px;min-width: 200px">surprised</td>
576                        <td style="text-align: center"><audio controls="controls" style="width: 160px;"><source src="audios/ravdess/2_prompt_8.wav">Your browser does not support the audio element.</audio></td>
577                        <td style="text-align: center"><audio controls="controls" style="width: 160px;"><source src="audios/ravdess/2_gt_8.wav">Your browser does not support the audio element.</audio></td>
578                        <td style="text-align: center"><audio controls="controls" style="width: 160px;"><source src="audios/ravdess/gt/rvd_bm_7_surprised_ours.wav">Your browser does not support the audio element.</audio></td>
579                        <td style="text-align: center"><audio controls="controls" style="width: 160px;"><source src="audios/ravdess/ns3/rvd_bm_7_surprised.wav">Your browser does not support the audio element.</audio></td>
580                        <td style="text-align: center"><audio controls="controls" style="width: 160px;"><source src="audios/ravdess/ns2/rvd_bm_7_surprised.wav">Your browser does not support the audio element.</audio></td>
581                        <td style="text-align: center"><audio controls="controls" style="width: 160px;"><source src="audios/ravdess/voiceboxr/rvd_bm_7_surprised.wav">Your browser does not support the audio element.</audio></td>
582                        <td style="text-align: center"><audio controls="controls" style="width: 160px;"><source src="audios/ravdess/valler/rvd_bm_7_surprised.wav">Your browser does not support the audio element.</audio></td>
583                        <td style="text-align: center"><audio controls="controls" style="width: 160px;"><source src="audios/ravdess/megatts2/rvd_bm_7_surprised.wav">Your browser does not support the audio element.</audio></td>
584                        <td style="text-align: center"><audio controls="controls" style="width: 160px;"><source src="audios/ravdess/styletts2/rvd_bm_7_surprised.wav">Your browser does not support the audio element.</audio></td>
585                        <td style="text-align: center"><audio controls="controls" style="width: 160px;"><source src="audios/ravdess/hierspeechpp/rvd_bm_7_surprised.wav">Your browser does not support the audio element.</audio></td>
586                      </tr>
587                
588                    </tbody>
589                  </table>
590                  </div>
591                </div>
592              </div>
593              <br>
594              <br>
595              <h2 id="robustness" style="text-align: center">Zero-Shot TTS Samples (Emotion)</h2>
596              <p>
597                The text “Why fades the lotus of the water” is extracted from the LibriSpeech dataset.
598                Baseline samples are taken from the NaturalSpeech 3 demo<sup
599                  ><a href="#footnote5-3">3</a></sup
600                >
601              </p>
602              <div class="card card-body ditto-card">
603              <div class="ditto-sample-box">
604                <div class="table-responsive pt-3">
605                  <table class="table table-hover pt-2">
606                    <thead>
607                      <tr>
608                      <th style="text-align: center">Prompt Emotion</th>
609                      <th style="text-align: center">Prompt</th>
610                      <th style="text-align: center">DiTTo-TTS</th>
611                      <th style="text-align: center">NaturalSpeech 3</th>
612                      </tr>
613                    </thead>
614                    <tbody>
615                      <tr>
616                        <td style="text-align: center;vertical-align:middle;width: 200px;min-width: 200px">neutral</td>
617                        <td style="text-align: center"><audio controls="controls" style="width: 200px;"><source src="audios/ravdess/2_prompt_9.wav">Your browser does not support the audio element.</audio></td>
618                        <td style="text-align: center"><audio controls="controls" style="width: 200px;"><source src="audios/ravdess/zs_emo/zs_emo_0_neutral_ours.wav">Your browser does not support the audio element.</audio></td>
619                        <td style="text-align: center"><audio controls="controls" style="width: 200px;"><source src="audios/ravdess/ns3_zs_emo/zs_emo_0_neutral.wav">Your browser does not support the audio element.</audio></td>
620                      </tr>
621                
622                      <tr>
623                        <td style="text-align: center;vertical-align:middle;width: 200px;min-width: 200px">happy</td>
624                        <td style="text-align: center"><audio controls="controls" style="width: 200px;"><source src="audios/ravdess/2_prompt_10.wav">Your browser does not support the audio element.</audio></td>
625                        <td style="text-align: center"><audio controls="controls" style="width: 200px;"><source src="audios/ravdess/zs_emo/zs_emo_1_happy_ours.wav">Your browser does not support the audio element.</audio></td>
626                        <td style="text-align: center"><audio controls="controls" style="width: 200px;"><source src="audios/ravdess/ns3_zs_emo/zs_emo_1_happy.wav">Your browser does not support the audio element.</audio></td>
627                      </tr>
628                
629                      <tr>
630                        <td style="text-align: center;vertical-align:middle;width: 200px;min-width: 200px">calm</td>
631                        <td style="text-align: center"><audio controls="controls" style="width: 200px;"><source src="audios/ravdess/2_prompt_11.wav">Your browser does not support the audio element.</audio></td>
632                        <td style="text-align: center"><audio controls="controls" style="width: 200px;"><source src="audios/ravdess/zs_emo/zs_emo_2_calm_ours.wav">Your browser does not support the audio element.</audio></td>
633                        <td style="text-align: center"><audio controls="controls" style="width: 200px;"><source src="audios/ravdess/ns3_zs_emo/zs_emo_2_calm.wav">Your browser does not support the audio element.</audio></td>
634                      </tr>
635                
636                      <tr>
637                        <td style="text-align: center;vertical-align:middle;width: 200px;min-width: 200px">sad</td>
638                        <td style="text-align: center"><audio controls="controls" style="width: 200px;"><source src="audios/ravdess/2_prompt_12.wav">Your browser does not support the audio element.</audio></td>
639                        <td style="text-align: center"><audio controls="controls" style="width: 200px;"><source src="audios/ravdess/zs_emo/zs_emo_3_sad_ours.wav">Your browser does not support the audio element.</audio></td>
640                        <td style="text-align: center"><audio controls="controls" style="width: 200px;"><source src="audios/ravdess/ns3_zs_emo/zs_emo_3_sad.wav">Your browser does not support the audio element.</audio></td>
641                      </tr>
642                
643                      <tr>
644                        <td style="text-align: center;vertical-align:middle;width: 200px;min-width: 200px">angry</td>
645                        <td style="text-align: center"><audio controls="controls" style="width: 200px;"><source src="audios/ravdess/2_prompt_13.wav">Your browser does not support the audio element.</audio></td>
646                        <td style="text-align: center"><audio controls="controls" styl
646e="width: 200px;"><source src="audios/ravdess/zs_emo/zs_emo_4_angry_ours.wav">Your browser does not support the audio element.</audio></td>
647                        <td style="text-align: center"><audio controls="controls" style="width: 200px;"><source src="audios/ravdess/ns3_zs_emo/zs_emo_4_angry.wav">Your browser does not support the audio element.</audio></td>
648                      </tr>
649                
650                      <tr>
651                        <td style="text-align: center;vertical-align:middle;width: 200px;min-width: 200px">fearful</td>
652                        <td style="text-align: center"><audio controls="controls" style="width: 200px;"><source src="audios/ravdess/2_prompt_14.wav">Your browser does not support the audio element.</audio></td>
653                        <td style="text-align: center"><audio controls="controls" style="width: 200px;"><source src="audios/ravdess/zs_emo/zs_emo_5_fearful_ours.wav">Your browser does not support the audio element.</audio></td>
654                        <td style="text-align: center"><audio controls="controls" style="width: 200px;"><source src="audios/ravdess/ns3_zs_emo/zs_emo_5_fearful.wav">Your browser does not support the audio element.</audio></td>
655                      </tr>
656                
657                      <tr>
658                        <td style="text-align: center;vertical-align:middle;width: 200px;min-width: 200px">disgust</td>
659                        <td style="text-align: center"><audio controls="controls" style="width: 200px;"><source src="audios/ravdess/2_prompt_15.wav">Your browser does not support the audio element.</audio></td>
660                        <td style="text-align: center"><audio controls="controls" style="width: 200px;"><source src="audios/ravdess/zs_emo/zs_emo_6_disgust_ours.wav">Your browser does not support the audio element.</audio></td>
661                        <td style="text-align: center"><audio controls="controls" style="width: 200px;"><source src="audios/ravdess/ns3_zs_emo/zs_emo_6_disgust.wav">Your browser does not support the audio element.</audio></td>
662                      </tr>
663                
664                      <tr>
665                        <td style="text-align: center;vertical-align:middle;width: 200px;min-width: 200px">surprised</td>
666                        <td style="text-align: center"><audio controls="controls" style="width: 200px;"><source src="audios/ravdess/2_prompt_16.wav">Your browser does not support the audio element.</audio></td>
667                        <td style="text-align: center"><audio controls="controls" style="width: 200px;"><source src="audios/ravdess/zs_emo/zs_emo_7_surprised_ours.wav">Your browser does not support the audio element.</audio></td>
668                        <td style="text-align: center"><audio controls="controls" style="width: 200px;"><source src="audios/ravdess/ns3_zs_emo/zs_emo_7_surprised.wav">Your browser does not support the audio element.</audio></td>
669                      </tr>
670                    </tbody>
671                  </table>
672                  </div>
673                </div>
674              </div>
675
676              <sup id="footnote5-1">1</sup
677              ><a href="https://mega-tts.github.io/demo-page/"
678                >https://mega-tts.github.io/demo-page/</a
679              ><br>
680              <sup id="footnote5-2">2</sup
681              ><a href="https://clam-tts.github.io/"
682                >https://clam-tts.github.io/</a
683              ><br>
684              <sup id="footnote5-3">3</sup
685              ><a href="https://speechresearch.github.io/naturalspeech3/"
686                >https://speechresearch.github.io/naturalspeech3/</a
687              >
688            </div>
689
690            <div
691              id="speech-diversity-box"
692              role="tabpanel"
693              class="tab-pane fade container pt-5 shadow p-5 mb-5 bg-white rounded"
694            >
695              <h2 id="speech-diversity" style="text-align: center">
696                Speech Diversity
697              </h2>
698              <p>
699                DiTTo-TTS can synthesize speech with diverse <b>prosody and timbre</b>,
700                and even <b>environmental background sounds</b>, from <b>various speakers</b>
701                when generated without a prompt. First three text scripts are from Gigaspeech
702                test set, and the last four are brought from
703                Vall-E demo<sup><a href="#footnote6-1">1</a></sup>.
704              </p>
705              <div class="card card-body ditto-card">
706                <div class="ditto-sample-box">
707                  <div class="table-responsive pt-3">
708                    <table class="table table-hover pt-2">
709                      <thead>
710                        <tr>
711                          <th>Text</th>
712                          <th>Sample #1</th>
713                          <th>Sample #2</th>
714                          <th>Sample #3</th>
715                          <th>Sample #4</th>
716                          <th>Sample #5</th>
717                        </tr>
718                      </thead>
719                      <tbody>
720                        <tr>
721                          <td data-label="Text">
722                            so it’s really for us. it’s kind of like the ah
723                            political, social headquarters or center for our
724                            community.
725                          </td>
726                          <td data-label="Sample #1">
727                            <audio
728                              controls="controls"
729                              preload="none"
730                              src="audios/speaker_diversity/so_it_000.wav"
731                            ></audio>
732                          </td>
733                          <td data-label="Sample #2">
734                            <audio
735                              controls="controls"
736                              preload="none"
737                              src="audios/speaker_diversity/so_it_001.wav"
738                            ></audio>
739                          </td>
740                          <td data-label="Sample #3">
741                            <audio
742                              controls="controls"
743                              preload="none"
744                              src="audios/speaker_diversity/so_it_002.wav"
745                            ></audio>
746                          </td>
747                          <td data-label="Sample #4">
748                            <audio
749                              controls="controls"
750                              preload="none"
751                              src="audios/speaker_diversity/so_it_003.wav"
752                            ></audio>
753                          </td>
754                          <td data-label="Sample #5">
755                            <audio
756                              controls="controls"
757                              preload="none"
758                              src="audios/speaker_diversity/so_it_004.wav"
759                            ></audio>
760                          </td>
761                        </tr>
762                        <tr>
763                          <td data-label="Text">
764                            provided it’s evidence evidence-based and rigorous
765                            but, also, i believe, fair.
766                          </td>
767                          <td data-label="Sample #1">
768                            <audio
769                              controls="controls"
770                              preload="none"
771                              src="audios/speaker_diversity/provi_000.wav"
772                            ></audio>
773                          </td>
774                          <td data-label="Sample #2">
775                            <audio
776                              controls="controls"
777                              preload="none"
778                              src="audios/speaker_diversity/provi_001.wav"
779                            ></audio>
780                          </td>
781                          <td data-label="Sample #3">
782                            <audio
783                              controls="controls"
784                              preload="none"
785                              src="audios/speaker_diversity/provi_002.wav"
786                            ></audio>
787                          </td>
788                          <td data-label="Sample #4">
789                            <audio
790                              controls="controls"
791                              preload="none"
792                              src="audios/speaker_diversity/provi_003.wav"
793                            ></audio>
794                          </td>
795                          <td data-label="Sample #5">
796                            <audio
797                              controls="controls"
798                              preload="none"
799                              src="audios/speaker_diversity/provi_004.wav"
800                            ></audio>
801                          </td>
802                        </tr>
803                        <tr>
804                          <td data-label="Text">
805                            now you you've been with the office in Northern
806                            Ireland for ten years.
807                          </td>
808                          <td data-label="Sample #1">
809                            <audio
810                              controls="controls"
811                              preload="none"
812                              src="audios/speaker_diversity/now_000.wav"
813                            ></audio>
814                          </td>
815                          <td data-label="Sample #2">
816                            <audio
817                              controls="controls"
818                              preload="none"
819                              src="audios/speaker_diversity/now_001.wav"
820                            ></audio>
821                          </td>
822                          <td data-label="Sample #3">
823                            <audio
824                              controls="controls"
825                              preload="none"
826                              src="audios/speaker_diversity/now_002.wav"
827                            ></audio>
828                          </td>
829                          <td data-label="Sample #4">
830                            <audio
831                              controls="controls"
832                              preload="none"
833                              src="audios/speaker_diversity/now_003.wav"
834                            ></audio>
835                          </td>
836                          <td data-label="Sample #5">
837                            <audio
838                              controls="controls"
839                              preload="none"
840                              src="audios/speaker_diversity/now_004.wav"
841                            ></audio>
842                          </td>
843                        </tr>
844                        <tr>
845                          <td data-label="Text">
846                            Number ten, fresh nelly is waiting on you, good
847                            night husband.
848                          </td>
849                          <td data-label="Sample #1">
850                            <audio
851                              controls="controls"
852                              preload="none"
853                              src="audios/speaker_diversity/Numbe_000.wav"
854                            ></audio>
855                          </td>
856                          <td data-label="Sample #2">
857                            <audio
858                              controls="controls"
859                              preload="none"
860                              src="audios/speaker_diversity/Numbe_001.wav"
861                            ></audio>
862                          </td>
863                          <td data-label="Sample #3">
864                            <audio
865                              controls="controls"
866                              preload="none"
867                              src="audios/speaker_diversity/Numbe_002.wav"
868                            ></audio>
869                          </td>
870                          <td data-label="Sample #4">
871                            <audio
872                              controls="controls"
873                              preload="none"
874                              src="audios/speaker_diversity/Numbe_003.wav"
875                            ></audio>
876                          </td>
877                          <td data-label="Sample #5">
878                            <audio
879                              controls="controls"
880                              preload="none"
881                              src="audios/speaker_diversity/Numbe_004.wav"
882                            ></audio>
883                          </td>
884                        </tr>
885                        <tr>
886                          <td data-label="Text">Because we do not need it.</td>
887                          <td data-label="Sample #1">
888                            <audio
889                              controls="controls"
890                              preload="none"
891                              src="audios/speaker_diversity/Becau_000.wav"
892                            ></audio>
893                          </td>
894                          <td data-label="Sample #2">
895                            <audio
896                              controls="controls"
897                              preload="none"
898                              src="audios/speaker_diversity/Becau_001.wav"
899                            ></audio>
900                          </td>
901                          <td data-label="Sample #3">
902                            <audio
903                              controls="controls"
904                              preload="none"
905                              src="audios/speaker_diversity/Becau_002.wav"
906                            ></audio>
907                          </td>
908                          <td data-label="Sample #4">
909                            <audio
910                              controls="controls"
911                              preload="none"
912                              src="audios/speaker_diversity/Becau_003.wav"
913                            ></audio>
914                          </td>
915                          <td data-label="Sample #5">
916                            <audio
917                              controls="controls"
918                              preload="none"
919                              src="audios/speaker_diversity/Becau_004.wav"
920                            ></audio>
921                          </td>
922                        </tr>
923                        <tr>
924                          <td data-label="Text">
925                            I must do something about it.
926                          </td>
927                          <td data-label="Sample #1">
928                            <audio
929                              controls="controls"
930                              preload="none"
931                              src="audios/speaker_diversity/I_mus_000.wav"
932                            ></audio>
933                          </td>
934                          <td data-label="Sample #2">
935                            <audio
936                              controls="controls"
937                              preload="none"
938                              src="audios/speaker_diversity/I_mus_001.wav"
939                            ></audio>
940                          </td>
941                          <td data-label="Sample #3">
942                            <audio
943                              controls="controls"
944                              preload="none"
945                              src="audios/speaker_diversity/I_mus_002.wav"
946                            ></audio>
947                          </td>
948                          <td data-label="Sample #4">
949                            <audio
950                              controls="controls"
951                              preload="none"
952                              src="audios/speaker_diversity/I_mus_003.wav"
953                            ></audio>
954                          </td>
955                          <td data-label="Sample #5">
956                            <audio
957                              controls="controls"
958                              preload="none"
959                              src="audios/speaker_diversity/I_mus_004.wav"
960                            ></audio>
961                          </td>
962                        </tr>
963                        <tr>
964                          <td data-label="Text">He has not been named.</td>
965                          <td data-label="Sample #1">
966                            <audio
967                              controls="controls"
968                              preload="none"
969                              src="audios/speaker_diversity/He_ha_000.wav"
970                            ></audio>
971                          </td>
972                          <td data-label="Sample #2">
973                            <audio
974                              controls="controls"
975                              preload="none"
976                              src="audios/speaker_diversity/He_ha_001.wav"
977                            ></audio>
978                          </td>
979                          <td data-label="Sample #3">
980                            <audio
981                              controls="controls"
982                              preload="none"
983                              src="audios/speaker_diversity/He_ha_002.wav"
984                            ></audio>
985                          </td>
986                          <td data-label="Sample #4">
987                            <audio
988                              controls="controls"
989                              preload="none"
990                              src="audios/speaker_diversity/He_ha_003.wav"
991                            ></audio>
992                          </td>
993                          <td data-label="Sample #5">
994                            <audio
995                              controls="controls"
996                              preload="none"
997                              src="audios/speaker_diversity/He_ha_004.wav"
998                            ></audio>
999                          </td>
1000                        </tr>
1001                      </tbody>
1002                    </table>
1003                  </div>
1004                </div>
1005              </div>
1006              <sup id="footnote6-1">1</sup
1007              ><a
1008                href="https://www.microsoft.com/en-us/research/project/vall-e-x/vall-e/"
1009                >https://www.microsoft.com/en-us/research/project/vall-e-x/vall-e/</a
1010              >
1011            </div>
1012            <div
1013              id="speech-rate-contrl-box"
1014              role="tabpanel"
1015              class="tab-pane fade container pt-5 shadow p-5 mb-5 bg-white rounded"
1016            >
1017              <h2 id="speech-rate-contrl" style="text-align: center">Speech Rate Controllability</h2>
1018              <p>
1019                DiTTo-TTS can synthesize the same text into audio of varying lengths.
1020                This variable-length modeling allows for speech rate control by adjusting
1021                the total duration of the generated speech, as shown in the following
1022                mel-spectrogram. The speech rate decreases from top to bottom.
1023                For more details, refer to <b>Section 5.2</b> of our paper.
1024              </p>
1025              <p style="text-align: center">
1026                <img src="asset/speech_rate_control.png" height="400" width="600"
1027                  class="img-fluid" />
1028              </p>
1029              <div class="card card-body ditto-card">
1030                <div class="ditto-sample-box">
1031                  <div class="table-responsive pt-3">
1032                    <table class="table table-hover pt-2">
1033                      <thead>
1034                        <tr>
1035                          <th>Text</th>
1036                          <th>Length of Latent Audio</th>
1037                          <th>Sample</th>
1038                        </tr>
1039                      </thead>
1040                      <tbody>
1041                        <tr>
1042                          <td data-label="Text" rowspan="4">But to those who knew her well, it was a symbol of <br> her unwavering determination and spirit.</td>
1043                          <td data-label="Length">38</td>
1044                          <td data-label="Sample">
1045                            <audio controls="controls" preload="none" src="audios/speech_rate/speech_rate_control_0.wav"></audio>
1046                          </td>
1047                        </tr>
1048                        <tr>
1049                          <td data-label="Length">43</td>
1050                          <td data-label="Sample">
1051                            <audio controls="controls" preload="none" src="audios/speech_rate/speech_rate_control_1.wav"></audio>
1052                          </td>
1053                        </tr>
1054                        <tr>
1055                          <td data-label="Length">53</td>
1056                          <td data-label="Sample">
1057                            <audio controls="controls" preload="none" src="audios/speech_rate/speech_rate_control_2.wav"></audio>
1058                          </td>
1059                        </tr>
1060                        <tr>
1061                          <td data-label="Length">63</td>
1062                          <td data-label="Sample">
1063                            <audio controls="controls" preload="none" src="audios/speech_rate/speech_rate_control_3.wav"></audio>
1064                          </td>
1065                        </tr>
1066                      </tbody>
1067                    </table>
1068                  </div>
1069                </div>
1070              </div>
1071            </div>
1072            <div
1073              id="speech-editing-box"
1074              role="tabpanel"
1075              class="tab-pane fade container pt-5 shadow p-5 mb-5 bg-white rounded"
1076            >
1077              <h2 id="speech-editing" style="text-align: center">
1078                Speech Editing
1079              </h2>
1080              <p>
1081                DiTTo-TTS is capable of correcting mispronounced words by editing generated speech contents,
1082                without the need for the speaker to re-record the audio. The examples are brought from
1083                Voicebox demo<sup><a href="#footnote7-1">1</a></sup> and
1084                VoiceCraft demo<sup><a href="#footnote7-2">2</a></sup>.
1085              </p>
1086              <div class="card card-body ditto-card">
1087                <div class="ditto-sample-box">
1088                  <div class="table-responsive pt-3">
1089                    <table class="table table-hover pt-2">
1090                      <thead>
1091                        <tr>
1092                          <th>Original Text</th>
1093                          <th>Original Audio</th>
1094                          <th>Edited Text</th>
1095                          <th>Voicebox<br>(16 kHz)</th>
1096                          <th>VoiceCraft<br>(16 kHz)</th>
1097                          <th>DiTTo-TTS<br>(22.05 kHz)</th>
1098                        </tr>
1099                      </thead>
1100                      <tbody>
1101                        <tr>
1102                          <td data-label="Original Text">
1103                            in zero weather in mid-winter when <b>the earth is frozen to a great depth below the surface</b> when in driving over the unpaved country roads they give forth a hard metallic road
1104                          </td>
1105                          <td data-label="Original Audio">
1106                            <audio
1107                            controls="controls"
1108                            preload="none"
1109                            src="audios/speech_editing/0.wav"
1110                            ></audio>
1111                          </td>
1112                          <td data-label="Edited Text">
1113                            in zero weather in mid-winter when <b>jack frost has cast his icy spell upon the land</b> when in driving over the unpaved country roads they give forth a hard metallic road
1114                          </td>
1115                          <td data-label="Voicebox (16 kHz)">
1116                            <audio
1117                              controls="controls"
1118                              preload="none"
1119                              src="audios/speech_editing/0_0.wav"
1120                            ></audio>
1121                          </td>
1122                          <td data-label="VoiceCraft (16 kHz)">
1123                            <audio
1124                              controls="controls"
1125                              preload="none"
1126                              src="audios/speech_editing/0_0_voicecraft.wav"
1127                            ></audio>
1128                          </td>
1129                          <td data-label="DiTTo-TTS (22.05 kHz)">
1130                            <audio
1131                              controls="controls"
1132                              preload="none"
1133                              src="audios/speech_editing/0_ditto_all.wav"
1134                            ></audio>
1135                          </td>
1136                        </tr>
1137                        <tr>
1138                          <td data-label="Original Text">
1139                            and especially as i am not very much up in latin myself he said the suit was on <b>an insurance policy</b> that he was defending on the ground of misinterpretations
1140                          </td>
1141                          <td data-label="Original Audio">
1142                            <audio
1143                            controls="controls"
1144                            preload="none"
1145                            src="audios/speech_editing/1.wav"
1146                            ></audio>
1147                          </td>
1148                          <td data-label="Edited Text">
1149                            and especially as i am not very much up in latin myself he said the suit was on <b>a classified treasure map</b> that he was defending on the ground of misinterpretations
1150                          </td>
1151                          <td data-label="Voicebox (16 kHz)">
1152                            <audio
1153                              controls="controls"
1154                              preload="none"
1155                              src="audios/speech_editing/1_0.wav"
1156                            ></audio>
1157                          </td>
1158                          <td data-label="VoiceCraft (16 kHz)">
1159                            <audio
1160                              controls="controls"
1161                              preload="none"
1162                              src="audios/speech_editing/1_0_voicecraft.wav"
1163                            ></audio>
1164                          </td>
1165                          <td data-label="DiTTo-TTS (22.05 kHz)">
1166                            <audio
1167                              controls="controls"
1168                              preload="none"
1169                              src="audios/speech_editing/1_ditto_all.wav"
1170                            ></audio>
1171                          </td>
1172                        </tr>
1173                        <tr>
1174                          <td data-label="Original Text">
1175                            yet these petty operations incessantly continued in time surmount the greatest difficulties <b>mountains are elevated and oceans bounded</b> by the slender force of human beings
1176                          </td>
1177                          <td data-label="Original Audio">
1178                            <audio
1179                            controls="controls"
1180                            preload="none"
1181                            src="audios/speech_editing/2.wav"
1182                            ></audio>
1183                          </td>
1184                          <td data-label="Edited Text">
1185                            yet these petty operations incessantly continued in time surmount the greatest difficulties <b>vast challenges emerge and unexplored frontiers beckon</b> by the slender force of human beings
1186                          </td>
1187                          <td data-label="Voicebox (16 kHz)">
1188                            <audio
1189                              controls="controls"
1190                              preload="none"
1191                              src="audios/speech_editing/2_0.wav"
1192                            ></audio>
1193                          </td>
1194                          <td data-label="VoiceCraft (16 kHz)">
1195                            <audio
1196                              controls="controls"
1197                              preload="none"
1198                              src="audios/speech_editing/2_0_voicecraft.wav"
1199                            ></audio>
1200                          </td>
1201                          <td data-label="DiTTo-TTS (22.05 kHz)">
1202                            <audio
1203                              controls="controls"
1204                              preload="none"
1205                              src="audios/speech_editing/2_ditto_all.wav"
1206                            ></audio>
1207                          </td>
1208                        </tr>
1209                        <tr>
1210                          <td data-label="Original Text">
1211                            will find himself completely at a loss on <b>occasions of common and constant recurrence</b> speculative ability is one thing and practical ability is another
1212                          </td>
1213                          <td data-label="Original Audio">
1214                            <audio
1215                            controls="controls"
1216                            preload="none"
1217                            src="audios/speech_editing/3.wav"
1218                            ></audio>
1219                          </td>
1220                          <td data-label="Edited Text">
1221                            will find himself completely at a loss on <b>rare and unpredictable circumstances</b> speculative ability is one thing and practical ability is another
1222                          </td>
1223                          <td data-label="Voicebox (16 kHz)">
1224                            <audio
1225                              controls="controls"
1226                              preload="none"
1227                              src="audios/speech_editing/3_0.wav"
1228                            ></audio>
1229                          </td>
1230                          <td data-label="VoiceCraft (16 kHz)">
1231                            <audio
1232                              controls="controls"
1233                              preload="none"
1234                              src="audios/speech_editing/3_0_voicecraft.wav"
1235                            ></audio>
1236                          </td>
1237                          <td data-label="DiTTo-TTS (22.05 kHz)">
1238                            <audio
1239                              controls="controls"
1240                              preload="none"
1241                              src="audios/speech_editing/3_ditto_all.wav"
1242                            ></audio>
1243                          </td>
1244                        </tr>
1245                        <tr>
1246                          <td data-label="Original Text">
1247                            and <b>the carlsruhe</b> professor had to devise an ingenious apparatus which enabled him to bring the preparation at the required temperature on to the very plate of the microscope
1248                          </td>
1249                          <td data-label="Original Audio">
1250                            <audio
1251                            controls="controls"
1252                            preload="none"
1253                            src="audios/speech_editing/4.wav"
1254                            ></audio>
1255                          </td>
1256                          <td data-label="Edited Text">
1257                            and <b>the inventive</b> professor had to devise an ingenious apparatus which enabled him to bring the preparation at the required temperature on to the very plate of the microscope
1258                          </td>
1259                          <td data-label="Voicebox (16 kHz)">
1260                            <audio
1261                              controls="controls"
1262                              preload="none"
1263                              src="audios/speech_editing/4_0.wav"
1264                            ></audio>
1265                          </td>
1266                          <td data-label="VoiceCraft (16 kHz)">
1267                            <audio
1268                              controls="controls"
1269                              preload="none"
1270                              src="audios/speech_editing/4_0_voicecraft.wav"
1271                            ></audio>
1272                          </td>
1273                          <td data-label="DiTTo-TTS (22.05 kHz)">
1274                            <audio
1275                              controls="controls"
1276                              preload="none"
1277                              src="audios/speech_editing/4_ditto_all.wav"
1278                            ></audio>
1279                          </td>
1280                        </tr>
1281                        <tr>
1282                          <td data-label="Original Text">
1283                            this was george steers the son of a british naval captain and ship modeler who had become an american naval officer and was <b>the first man to take charge of the washington navy yard</b>
1284                          </td>
1285                          <td data-label="Original Audio">
1286                            <audio
1287                            controls="controls"
1288                            preload="none"
1289                            src="audios/speech_editing/5.wav"
1290                            ></audio>
1291                          </td>
1292                          <td data-label="Edited Text">
1293                            this was george steers the son of a british naval captain and ship modeler who had become an american naval officer and was <b>entrusted with the prestigious role of overseeing the operations at the renowned naval headquarters</b>
1294                          </td>
1295                          <td data-label="Voicebox (16 kHz)">
1296                            <audio
1297                              controls="controls"
1298                              preload="none"
1299                              src="audios/speech_editing/5_0.wav"
1300                            ></audio>
1301                          </td>
1302                          <td data-label="VoiceCraft (16 kHz)">
1303                            <audio
1304                              controls="controls"
1305                              preload="none"
1306                              src="audios/speech_editing/5_0_voicecraft.wav"
1307                            ></audio>
1308                          </td>
1309                          <td data-label="DiTTo-TTS (22.05 kHz)">
1310                            <audio
1311                              controls="controls"
1312                              preload="none"
1313                              src="audio
1313s/speech_editing/5_ditto_all.wav"
1314                            ></audio>
1315                          </td>
1316                        </tr>
1317                      </tbody>
1318                    </table>
1319                  </div>
1320                </div>
1321              </div>
1322              <sup id="footnote7-1">1</sup
1323              ><a
1324                href="https://voicebox.metademolab.com/edit.html"
1325                >https://voicebox.metademolab.com/edit.html</a
1326              ><br>
1327              <sup id="footnote7-2">2</sup
1328              ><a
1329                href="https://jasonppy.github.io/VoiceCraft_web/"
1330                >https://jasonppy.github.io/VoiceCraft_web/</a
1331              >
1332            </div>
1333            <div
1334              id="compare-box"
1335              role="tabpanel"
1336              class="tab-pane fade container pt-5 shadow p-5 mb-5 bg-white rounded"
1337            >
1338              <h2 id="compare" style="text-align: center">Comparison with Zero-Shot Samples and Demo Cases</h2>
1339              <p>
1340                We provide some comparison samples to compare DiTTo-TTS with four baseline models.
1341                To generate samples from Simple-TTS, we use
1342                <a href="https://github.com/asappresearch/simple-tts">the official checkpoints</a>
1343                and the text scripts from LibriSpeech test-clean subset. For the other three,
1344                E3-TTS, NaturalSpeech 3, and Voicebox, we download the audio samples from the
1345                demo pages of each baseline and then compare them with the audio generated by
1346                our model without any resampling. Demos for each model can be found at their respective links:
1347                <a href="https://e3tts.github.io/">E3-TTS</a>,
1348                <a href="https://speechresearch.github.io/naturalspeech3/">NaturalSpeech 3</a>, and
1349                <a href="https://voicebox.metademolab.com/zs_tts.html">Voicebox</a>.
1350                Please refer to <b>Section 5.1</b> and <b>Appendix A.3</b> of our
1351                paper for more details about subjective evaluation.
1352                The details of each baseline can be found in <b>Appendix A.4</b> of our paper.
1353              </p>
1354              <div class="table-responsive pt-3">
1355                <!-- <h3 style="text-align: center">Simple-TTS</h3> -->
1356                <table class="table table-hover pt-2">
1357                  <thead>
1358                    <tr>
1359                      <th style="text-align: center">Text</th>
1360                      <th style="text-align: center">Prompt</th>
1361                      <th style="text-align: center">Simple-TTS (24 kHz)</th>
1362                      <th style="text-align: center">DiTTo-TTS (22.05 kHz)</th>
1363                    </tr>
1364                  </thead>
1365                  <tbody>
1366                    <tr>
1367                      <td
1368                        style="text-align: left;vertical-align:middle;width: 300px">
1369                        do not therefore think that the gothic school is an easy one</td>
1370                      <td style="text-align: center"><audio controls="controls"
1371                          style="width: 300px;">
1372                          <source src="audios/comparison/simpletts/1188-133604-0014_wav_p_0_wav_p.wav" autoplay="">Your
1373                          browser
1374                          does not support the audio element.
1375                        </audio></td>
1376                      <td style="text-align: center"><audio controls="controls"
1377                          style="width: 300px;">
1378                          <source src="audios/comparison/simpletts/1188-133604-0014_wav_p_0.wav"
1379                            autoplay="">Your
1380                          browser does not support the audio element.
1381                        </audio></td>
1382                      <td style="text-align: center"><audio controls="controls"
1383                          style="width: 300px;">
1384                          <source
1385                            src="audios/comparison/simpletts/1188-133604-0014_wav_p_0_ours.wav"
1386                            autoplay="">Your
1387                          browser does not support the audio element.
1388                        </audio></td>
1389                    </tr>
1390                    <tr>
1391                      <td
1392                        style="text-align: left;vertical-align:middle;width: 300px">
1393                        did ever anybody see the like screamed missus poyser running towards the table when her eye had fallen on the blue stream</td>
1394                      <td style="text-align: center"><audio controls="controls"
1395                          style="width: 300px;">
1396                          <source src="audios/comparison/simpletts
1396/2094-142345-0029_wav_p_2_wav_p.wav" autoplay="">Your
1397                          browser
1398                          does not support the audio element.
1399                        </audio></td>
1400                      <td style="text-align: center"><audio controls="controls"
1401                          style="width: 300px;">
1402                          <source src="audios/comparison/simpletts/2094-142345-0029_wav_p_2.wav"
1403                            autoplay="">Your
1404                          browser does not support the audio element.
1405                        </audio></td>
1406                      <td style="text-align: center"><audio controls="controls"
1407                          style="width: 300px;">
1408                          <source
1409                            src="audios/comparison/simpletts/2094-142345-0029_wav_p_2_ours.wav"
1410                            autoplay="">Your
1411                          browser does not support the audio element.
1412                        </audio></td>
1413                    </tr>
1414                    <tr>
1415                      <td
1416                        style="text-align: left;vertical-align:middle;width: 300px">
1417                        it is sold everywhere but for the last three weeks nobody will use any snuff but that sold at the civet cat</td>
1418                      <td style="text-align: center"><audio controls="controls"
1419                          style="width: 300px;">
1420                          <source src="audios/comparison/simpletts/3729-6852-0029_wav_p_1_wav_p.wav" autoplay="">Your
1421                          browser
1422                          does not support the audio element.
1423                        </audio></td>
1424                      <td style="text-align: center"><audio controls="controls"
1425                          style="width: 300px;">
1426                          <source src="audios/comparison/simpletts/3729-6852-0029_wav_p_1.wav"
1427                            autoplay="">Your
1428                          browser does not support the audio element.
1429                        </audio></td>
1430                      <td style="text-align: center"><audio controls="controls"
1431                          style="width: 300px;">
1432                          <source
1433                            src="audios/comparison/simpletts/3729-6852-0029_wav_p_1_ours.wav"
1434                            autoplay="">Your
1435                          browser does not support the audio element.
1436                        </audio></td>
1437                    </tr>
1438                    <tr>
1439                      <td
1440                        style="text-align: left;vertical-align:middle;width: 300px">
1441                        and this was why kenneth and beth discovered him conversing with the young woman in the buggy</td>
1442                      <td style="text-align: center"><audio controls="controls"
1443                          style="width: 300px;">
1444                          <source src="audios/comparison/simpletts/6829-68771-0006_wav_p_1_wav_p.wav" autoplay="">Your
1445                          browser
1446                          does not support the audio element.
1447                        </audio></td>
1448                      <td style="text-align: center"><audio controls="controls"
1449                          style="width: 300px;">
1450                          <source src="audios/comparison/simpletts/6829-68771-0006_wav_p_1.wav"
1451                            autoplay="">Your
1452                          browser does not support the audio element.
1453                        </audio></td>
1454                      <td style="text-align: center"><audio controls="controls"
1455                          style="width: 300px;">
1456                          <source
1457                            src="audios/comparison/simpletts/6829-68771-0006_wav_p_1_ours.wav"
1458                            autoplay="">Your
1459                          browser does not support the audio element.
1460                        </audio></td>
1461                    </tr>
1462                  </tbody>
1463                </table>
1464              </div>
1465              <div class="table-responsive pt-3">
1466                <!-- <h3 style="text-align: center">E3-TTS</h3> -->
1467                <table class="table table-hover pt-2">
1468                  <thead>
1469                    <tr>
1470                      <th style="text-align: center">Text</th>
1471                      <th style="text-align: center">Prompt</th>
1472                      <th style="text-align: center">E3-TTS (24 kHz)</th>
1473                      <th style="text-align: center">DiTTo-TTS (22.05 kHz)</th>
1474                    </tr>
1475                  </thead>
1476                  <tbody>
1477                    <tr>
1478                      <td
1479                        style="text-align: left;vertical-align:middle;width: 300px">
1480                        She felt the force of the objections.</td>
1481                      <td style="text-align: center"><audio controls="controls"
1482                          style="width: 300px;">
1483                          <source src="audios/comparison/e3tts/prompt_02_wav_p.wav" autoplay="">Your
1484                          browser
1485                          does not support the audio element.
1486                        </audio></td>
1487                      <td style="text-align: center"><audio controls="controls"
1488                          style="width: 300px;">
1489                          <source src="audios/comparison/e3tts/prompt_02.wav"
1490                            autoplay="">Your
1491                          browser does not support the audio element.
1492                        </audio></td>
1493                      <td style="text-align: center"><audio controls="controls"
1494                          style="width: 300px;">
1495                          <source
1496                            src="audios/comparison/e3tts/prompt_02_ours.wav"
1497                            autoplay="">Your
1498                          browser does not support the audio element.
1499                        </audio></td>
1500                    </tr>
1501                    <tr>
1502                      <td
1503                        style="text-align: left;vertical-align:middle;width: 300px">
1504                        She can't get it out of her head, even after fifty years.</td>
1505                      <td style="text-align: center"><audio controls="controls"
1506                          style="width: 300px;">
1507                          <source src="audios/comparison/e3tts/prompt_03_wav_p.wav" autoplay="">Your
1508                          browser
1509                          does not support the audio element.
1510                        </audio></td>
1511                      <td style="text-align: center"><audio controls="controls"
1512                          style="width: 300px;">
1513                          <source src="audios/comparison/e3tts/prompt_03.wav"
1514                            autoplay="">Your
1515                          browser does not support the audio element.
1516                        </audio></td>
1517                      <td style="text-align: center"><audio controls="controls"
1518                          style="width: 300px;">
1519                          <source
1520                            src="audios/comparison/e3tts/prompt_03_ours.wav"
1521                            autoplay="">Your
1522                          browser does not support the audio element.
1523                        </audio></td>
1524                    </tr>
1525                    <tr>
1526                      <td
1527                        style="text-align: left;vertical-align:middle;width: 300px">
1528                        "I don't think so," replied Tom.</td>
1529                      <td style="text-align: center"><audio controls="controls"
1530                          style="width: 300px;">
1531                          <source src="audios/comparison/e3tts/prompt_13_wav_p.wav" autoplay="">Your
1532                          browser
1533                          does not support the audio element.
1534                        </audio></td>
1535                      <td style="text-align: center"><audio controls="controls"
1536                          style="width: 300px;">
1537                          <source src="audios/comparison/e3tts/prompt_13.wav"
1538                            autoplay="">Your
1539                          browser does not support the audio element.
1540                        </audio></td>
1541                      <td style="text-align: center"><audio controls="controls"
1542                          style="width: 300px;">
1543                          <source
1544                            src="audios/comparison/e3tts/prompt_13_ours.wav"
1545                            autoplay="">Your
1546                          browser does not support the audio element.
1547                        </audio></td>
1548                    </tr>
1549                    <tr>
1550                      <td
1551                        style="text-align: left;vertical-align:middle;width: 300px">
1552                        But at that moment the voice of the stranger was heard from the window.</td>
1553                      <td style="text-align: center"><audio controls="controls"
1554                          style="width: 300px;">
1555                          <source src="audios/comparison/e3tts/prompt_17_wav_p.wav" autoplay="">Your
1556                          browser
1557                          does not support the audio element.
1558                        </audio></td>
1559                      <td style="text-align: center"><audio controls="controls"
1560                          style="width: 300px;">
1561                          <source src="audios/comparison/e3tts/prompt_17.wav"
1562                            autoplay="">Your
1563                          browser does not support the audio element.
1564                        </audio></td>
1565                      <td style="text-align: center"><audio controls="controls"
1566                          style="width: 300px;">
1567                          <source
1568                            src="audios/comparison/e3tts/prompt_17_ours.wav"
1569                            autoplay="">Your
1570                          browser does not support the audio element.
1571                        </audio></td>
1572                    </tr>
1573                  </tbody>
1574                </table>
1575              </div>
1576              <div class="table-responsive pt-3">
1577                <!-- <h3 style="text-align: center">NaturalSpeech 3</h3> -->
1578                <table class="table table-hover pt-2">
1579                  <thead>
1580                    <tr>
1581                      <th style="text-align: center">Text</th>
1582                      <th style="text-align: center">Prompt</th>
1583                      <th style="text-align: center">NaturalSpeech 3 (16 kHz)</th>
1584                      <th style="text-align: center">DiTTo-TTS (22.05 kHz)</th>
1585                    </tr>
1586                  </thead>
1587                  <tbody>
1588                    <tr>
1589                      <td
1590                        style="text-align: left;vertical-align:middle;width: 300px">
1591                        They think you're proud because you've been away to school or something.</td>
1592                      <td style="text-align: center"><audio controls="controls"
1593                          style="width: 300px;">
1594                          <source src="audios/comparison/ns3/1_wav_p.wav" autoplay="">Your
1595                          browser
1596                          does not support the audio element.
1597                        </audio></td>
1598                      <td style="text-align: center"><audio controls="controls"
1599                          style="width: 300px;">
1600                          <source
1601                            src="audios/comparison/ns3/1.wav"
1602                            autoplay="">Your browser does not support the audio
1603                          element.
1604                        </audio></td>
1605                      <td style="text-align: center"><audio controls="controls"
1606                          style="width: 300px;">
1607                          <source
1608                            src="audios/comparison/ns3/1_ours.wav"
1609                            autoplay="">Your
1610                          browser does not support the audio
1611                          element.
1612                        </audio></td>
1613                    </tr>
1614                    <tr>
1615                      <td
1616                        style="text-align: left;vertical-align:middle;width: 300px">
1617                        There were only four stationers of any consequences in the town, and at each Holmes produced his pencil chips, and bid high for a duplicate.</td>
1618                      <td style="text-align: center"><audio controls="controls"
1619                          style="width: 300px;">
1620                          <source src="audios/comparison/ns3/4_wav_p.wav" autoplay="">Your
1621                          browser
1622                          does not support the audio element.
1623                        </audio></td>
1624                      <td style="text-align: center"><audio controls="controls"
1625                          style="width: 300px;">
1626                          <source
1627                            src="audios/comparison/ns3/4.wav"
1628                            autoplay="">Your browser does not support the audio
1629                          element.
1630                        </audio></td>
1631                      <td style="text-align: center"><audio controls="controls"
1632                          style="width: 300px;">
1633                          <source
1634                            src="audios/comparison/ns3/4_ours.wav"
1635                            autoplay="">Your
1636                          browser does not support the audio
1637                          element.
1638                        </audio></td>
1639                    </tr>
1640                    <tr>
1641                      <td
1642                        style="text-align: left;vertical-align:middle;width: 300px">
1643                        For, like as not, they must have thought him a prince when they saw his fine cap.</td>
1644                      <td style="text-align: center"><audio controls="controls"
1645                          style="width: 300px;">
1646                          <source src="audios/comparison/ns3/a_2_wav_p.wav" autoplay="">Your
1647                          browser
1648                          does not support the audio element.
1649                        </audio></td>
1650                      <td style="text-align: center"><audio controls="controls"
1651                          style="width: 300px;">
1652                          <source
1653                            src="audios/comparison/ns3/a_2.wav"
1654                            autoplay="">Your browser does not support the audio
1655                          element.
1656                        </audio></td>
1657                      <td style="text-align: center"><audio controls="controls"
1658                          style="width: 300px;">
1659                          <source
1660                            src="audios/comparison/ns3/a_2_ours.wav"
1661                            autoplay="">Your
1662                          browser does not support the audio
1663                          element.
1664                        </audio></td>
1665                    </tr>
1666                    <tr>
1667                      <td
1668                        style="text-align: left;vertical-align:middle;width: 300px">
1669                        The strong position held by the Edison system under the strenuous competition that was already springing up was enormously improved by the introduction of the three wire system and it gave an immediate impetus to incandescent lighting.</td>
1670                      <td style="text-align: center"><audio controls="controls"
1671                          style="width: 300px;">
1672                          <source src="audios/comparison/ns3/a_4_wav_p.wav" autoplay="">Your
1673                          browser
1674                          does not support the audio element.
1675                        </audio></td>
1676                      <td style="text-align: center"><audio controls="controls"
1677                          style="width: 300px;">
1678                          <source
1679                            src="audios/comparison/ns3/a_4.wav"
1680                            autoplay="">Your browser does not support the audio
1681                          element.
1682                        </audio></td>
1683                      <td style="text-align: center"><audio controls="controls"
1684                          style="width: 300px;">
1685                          <source
1686                            src="audios/comparison/ns3/a_4_ours.wav"
1687                            autoplay="">Your
1688                          browser does not support the audio
1689                          element.
1690                        </audio></td>
1691                    </tr>
1692                  </tbody>
1693                </table>
1694              </div>
1695              <div class="table-responsive pt-3">
1696                <!-- <h3 style="text-align: center">Voicebox</h3> -->
1697                <table class="table table-hover pt-2">
1698                  <thead>
1699                    <tr>
1700                      <th style="text-align: center">Text</th>
1701                      <th style="text-align: center">Prompt</th>
1702                      <th style="text-align: center">Voicebox (16 kHz)</th>
1703                      <th style="text-align: center">DiTTo-TTS (22.05 kHz)</th>
1704                    </tr>
1705                  </thead>
1706                  <tbody>
1707                    <tr>
1708                      <td
1709                        style="text-align: left;vertical-align:middle;width: 300px">
1710                        He was in deep converse with the clerk and entered the hall holding him by the arm</td>
1711                      <td style="text-align: center"><audio controls="controls"
1712                          style="width: 300px;">
1713                          <source src="audios/comparison/voicebox/61-70970-0007_wav_p.wav" autoplay="">Your
1714                          browser
1715                          does not support the audio element.
1716                        </audio></td>
1717                      <td style="text-align: center"><audio controls="controls"
1718                          style="width: 300px;">
1719                          <source
1720                            src="audios/comparison/voicebox/61-70970-0007.wav"
1721                            autoplay="">Your
1722                          browser does not support the audio
1723                          element.
1724                        </audio></td>
1725                      <td style="text-align: center"><audio controls="controls"
1726                          style="width: 300px;">
1727                          <source
1728                            src="audios/comparison/voicebox/61-70970-0007_ours.wav"
1729                            autoplay="">Your
1730                          browser does not support the audio
1731                          element.
1732                        </audio></td>
1733                    </tr>
1734                    <tr>
1735                      <td
1736                        style="text-align: left;vertical-align:middle;width: 300px">
1737                        Number ten fresh nelly is waiting on you good night husband</td>
1738                      <td style="text-align: center"><audio controls="controls"
1739                          style="width: 300px;">
1740                          <source src="audios/comparison/voicebox/1089-134686-0004_wav_p.wav" autoplay="">Your
1741                          browser
1742                          does not support the audio element.
1743                        </audio></td>
1744                      <td style="text-align: center"><audio controls="controls"
1745                          style="width: 300px;">
1746                          <source
1747                            src="audios/comparison/voicebox/1089-134686-0004.wav"
1748                            autoplay="">Your
1749                          browser does not support the audio
1750                          element.
1751                        </audio></td>
1752                      <td style="text-align: center"><audio controls="controls"
1753                          style="width: 300px;">
1754                          <source
1755                            src="audios/comparison/voicebox/1089-134686-0004_ours.wav"
1756                            autoplay="">Your
1757                          browser does not support the audio
1758                          element.
1759                        </audio></td>
1760                    </tr>
1761                    <tr>
1762                      <td
1763                        style="text-align: left;vertical-align:middle;width: 300px">
1764                        Yea his honourable worship is within but he hath a godly minister or tw
1764o with him and likewise a leech</td>
1765                      <td style="text-align: center"><audio controls="controls"
1766                          style="width: 300px;">
1767                          <source src="audios/comparison/voicebox/1221-135767-0014_wav_p.wav" autoplay="">Your
1768                          browser
1769                          does not support the audio element.
1770                        </audio></td>
1771                      <td style="text-align: center"><audio controls="controls"
1772                          style="width: 300px;">
1773                          <source
1774                            src="audios/comparison/voicebox/1221-135767-0014.wav"
1775                            autoplay="">Your
1776                          browser does not support the audio
1777                          element.
1778                        </audio></td>
1779                      <td style="text-align: center"><audio controls="controls"
1780                          style="width: 300px;">
1781                          <source
1782                            src="audios/comparison/voicebox/1221-135767-0014_ours.wav"
1783                            autoplay="">Your
1784                          browser does not support the audio
1785                          element.
1786                        </audio></td>
1787                    </tr>
1788                    <tr>
1789                      <td
1790                        style="text-align: left;vertical-align:middle;width: 300px">
1791                        Thus did this humane and right minded father comfort his unhappy daughter and her mother embracing her again did all she could to soothe her feelings</td>
1792                      <td style="text-align: center"><audio controls="controls"
1793                          style="width: 300px;">
1794                          <source src="audios/comparison/voicebox/5639-40744-0020_wav_p.wav" autoplay="">Your
1795                          browser
1796                          does not support the audio element.
1797                        </audio></td>
1798                      <td style="text-align: center"><audio controls="controls"
1799                          style="width: 300px;">
1800                          <source
1801                            src="audios/comparison/voicebox/5639-40744-0020.wav"
1802                            autoplay="">Your
1803                          browser does not support the audio
1804                          element.
1805                        </audio></td>
1806                      <td style="text-align: center"><audio controls="controls"
1807                          style="width: 300px;">
1808                          <source
1809                            src="audios/comparison/voicebox/5639-40744-0020_ours.wav"
1810                            autoplay="">Your
1811                          browser does not support the audio
1812                          element.
1813                        </audio></td>
1814                    </tr>
1815                  </tbody>
1816                </table>
1817              </div>
1818              <br>
1819              <br>
1820              <h2 id="compare" style="text-align: center">Comparison with LibriSpeech Zero-Shot Demo Cases</h2>
1821              <p>
1822                Here are some comparison samples to evaluate DiTTo-TTS against four other baseline models:
1823                Mega-TTS, NaturalSpeech 2, VALL-E, and CLaM-TTS. Our samples are downsampled to 16kHz for this section.
1824                Demos for each model can be found at their respective links:
1825                <a href="https://mega-tts.github.io/demo-page/">Mega-TTS</a>,
1826                <a href="https://speechresearch.github.io/naturalspeech2/">NaturalSpeech 2</a>,
1827                <a href="https://www.microsoft.com/en-us/research/project/vall-e-x/vall-e/">VALL-E</a>, and
1828                <a href="https://clam-tts.github.io/">CLaM-TTS</a>.
1829                The details of each baseline can be found in <b>Appendix A.4</b> of our paper.
1830              </p>
1831                  <div class="table-responsive pt-3">
1832                    <table class="table table-hover pt-2">
1833                    <thead>
1834                    <tr>
1835                    <th style="text-align: center">Text</th>
1836                    <th style="text-align: center">Prompt</th>
1837                    <th style="text-align: center">Ground Truth</th>
1838                    <th style="text-align: center">Mega-TTS</th>
1839                    <th style="text-align: center">NaturalSpeech 2</th>
1840                    <th style="text-align: center">VALL-E</th>
1841                    <th style="text-align: center">CLaM-TTS</th>
1842                    <th style="text-align: center">DiTTo-TTS</th>
1843                    </tr>
1844                    </thead>
1845                    <tbody>
1846                    <tr>
1847                    <td style="text-align: left;vertical-align:middle;width: 500px">They moved thereafter cautiously about the hut groping before and about them to find something to show that Warrenton had fulfilled his mission.</td>
1848                    <td style="text-align: center"><audio controls="controls" style="width: 160px;"><source src="audios/librispeech_16kHz/prompt/0.wav" autoplay="">Your browser does not support the audio element.</audio></td>
1849                    <td style="text-align: center"><audio controls="controls" style="width: 160px;"><source src="audios/librispeech_16kHz/gt/0.wav" autoplay="">Your browser does not support the audio element.</audio></td>
1850                    <td style="text-align: center; font-size: 18px; line-height: 2.5; padding-top: 10px; padding-bottom: 10px;">&mdash;</td>
1851                    <td style="text-align: center; font-size: 18px; line-height: 2.5; padding-top: 10px; padding-bottom: 10px;">&mdash;</td>
1852                    <td style="text-align: center"><audio controls="controls" style="width: 160px;"><source src="audios/librispeech_16kHz/valle/0.wav" autoplay="">Your browser does not support the audio element.</audio></td>
1853                    <td style="text-align: center"><audio controls="controls" style="width: 160px;"><source src="audios/librispeech_16kHz/clam/0.wav" autoplay="">Your browser does not support the audio element.</audio></td>
1854                    <td style="text-align: center"><audio controls="controls" style="width: 160px;"><source src="audios/librispeech_16kHz/ours/0.wav" autoplay="">Your browser does not support the audio element.</audio></td>
1855                    </tr>
1856                    <tr>
1857                      <td style="text-align: left;vertical-align:middle;width: 500px">And lay me down in thy cold bed and leave my shining lot.</td>
1858                      <td style="text-align: center"><audio controls="controls" style="width: 160px;"><source src="audios/librispeech_16kHz/prompt/1.wav" autoplay="">Your browser does not support the audio element.</audio></td>
1859                      <td style="text-align: center"><audio controls="controls" style="width: 160px;"><source src="audios/librispeech_16kHz/gt/1.wav" autoplay="">Your browser does not support the audio element.</audio></td>
1860                      <td style="text-align: center; font-size: 18px; line-height: 2.5; padding-top: 10px; padding-bottom: 10px;">&mdash;</td>
1861                      <td style="text-align: center"><audio controls="controls" style="width: 160px;"><source src="audios/librispeech_16kHz/ns2/1.wav" autoplay="">Your browser does not support the audio element.</audio></td>
1862                      <td style="text-align: center"><audio controls="controls" style="width: 160px;"><source src="audios/librispeech_16kHz/valle/1.wav" autoplay="">Your browser does not support the audio element.</audio></td>
1863                      <td style="text-align: center"><audio controls="controls" style="width: 160px;"><source src="audios/librispeech_16kHz/clam/1.wav" autoplay="">Your browser does not support the audio element.</audio></td>
1864                      <td style="text-align: center"><audio controls="controls" style="width: 160px;"><source src="audios/librispeech_16kHz/ours/1.wav" autoplay="">Your browser does not support the audio element.</audio></td>
1865                    </tr>
1866                    <tr>
1867                      <td style="text-align: left;vertical-align:middle;width: 500px">Number ten, fresh nelly is waiting on you, good night husband.</td>
1868                      <td style="text-align: center"><audio controls="controls" style="width: 160px;"><source src="audios/librispeech_16kHz/prompt/2.wav" autoplay="">Your browser does not support the audio element.</audio></td>
1869                      <td style="text-align: center"><audio controls="controls" style="width: 160px;"><source src="audios/librispeech_16kHz/gt/2.wav" autoplay="">Your browser does not support the audio element.</audio></td>
1870                      <td style="text-align: center; font-size: 18px; line-height: 2.5; padding-top: 10px; padding-bottom: 10px;">&mdash;</td>
1871                      <td style="text-align: center; font-size: 18px; line-height: 2.5; padding-top: 10px; padding-bottom: 10px;">&mdash;</td>
1872                      <td style="text-align: center"><audio controls="controls" style="width: 160px;"><source src="audios/librispeech_16kHz/valle/2.wav" autoplay="">Your browser does not support the audio element.</audio></td>
1873                      <td style="text-align: center"><audio controls="controls" style="width: 160px;"><source src="audios/librispeech_16kHz/clam/2.wav" autoplay="">Your browser does not support the audio element.</audio></td>
1874                      <td style="text-align: center"><audio controls="controls" style="width: 160px;"><source src="audios/librispeech_16kHz/ours/2.wav" autoplay="">Your browser does not support the audio element.</audio></td>
1875                    </tr>
1876                    <tr>
1877                      <td style="text-align: left;vertical-align:middle;width: 500px">Yea, his honourable worship is within, but he hath a godly minister or tw
1877o with him, and likewise a leech.</td>
1878                      <td style="text-align: center"><audio controls="controls" style="width: 160px;"><source src="audios/librispeech_16kHz/prompt/3.wav" autoplay="">Your browser does not support the audio element.</audio></td>
1879                      <td style="text-align: center"><audio controls="controls" style="width: 160px;"><source src="audios/librispeech_16kHz/gt/3.wav" autoplay="">Your browser does not support the audio element.</audio></td>
1880                      <td style="text-align: center"><audio controls="controls" style="width: 160px;"><source src="audios/librispeech_16kHz/megatts/3.wav" autoplay="">Your browser does not support the audio element.</audio></td>
1881                      <td style="text-align: center"><audio controls="controls" style="width: 160px;"><source src="audios/librispeech_16kHz/ns2/3.wav" autoplay="">Your browser does not support the audio element.</audio></td>
1882                      <td style="text-align: center"><audio controls="controls" style="width: 160px;"><source src="audios/librispeech_16kHz/valle/3.wav" autoplay="">Your browser does not support the audio element.</audio></td>
1883                      <td style="text-align: center"><audio controls="controls" style="width: 160px;"><source src="audios/librispeech_16kHz/clam/3.wav" autoplay="">Your browser does not support the audio element.</audio></td>
1884                      <td style="text-align: center"><audio controls="controls" style="width: 160px;"><source src="audios/librispeech_16kHz/ours/3.wav" autoplay="">Your browser does not support the audio element.</audio></td>
1885                    </tr>
1886                    <tr>
1887                      <td style="text-align: left;vertical-align:middle;width: 500px">Instead of shoes, the old man wore boots with turnover tops, and his blue coat had wide cuffs of gold braid.</td>
1888                      <td style="text-align: center"><audio controls="controls" style="width: 160px;"><source src="audios/librispeech_16kHz/prompt/4.wav" autoplay="">Your browser does not support the audio element.</audio></td>
1889                      <td style="text-align: center"><audio controls="controls" style="width: 160px;"><source src="audios/librispeech_16kHz/gt/4.wav" autoplay="">Your browser does not support the audio element.</audio></td>
1890                      <td style="text-align: center"><audio controls="controls" style="width: 160px;"><source src="audios/librispeech_16kHz/megatts/4.wav" autoplay="">Your browser does not support the audio element.</audio></td>
1891                      <td style="text-align: center; font-size: 18px; line-height: 2.5; padding-top: 10px; padding-bottom: 10px;">&mdash;</td>
1892                      <td style="text-align: center"><audio controls="controls" style="width: 160px;"><source src="audios/librispeech_16kHz/valle/4.wav" autoplay="">Your browser does not support the audio element.</audio></td>
1893                      <td style="text-align: center"><audio controls="controls" style="width: 160px;"><source src="audios/librispeech_16kHz/clam/4.wav" autoplay="">Your browser does not support the audio element.</audio></td>
1894                      <td style="text-align: center"><audio controls="controls" style="width: 160px;"><source src="audios/librispeech_16kHz/ours/4.wav" autoplay="">Your browser does not support the audio element.</audio></td>
1895                    </tr>
1896                    <tr>
1897                      <td style="text-align: left;vertical-align:middle;width: 500px">The army found the people in poverty and left them in comparative wealth.</td>
1898                      <td style="text-align: center"><audio controls="controls" style="width: 160px;"><source src="audios/librispeech_16kHz/prompt/5.wav" autoplay="">Your browser does not support the audio element.</audio></td>
1899                      <td style="text-align: center"><audio controls="controls" style="width: 160px;"><source src="audios/librispeech_16kHz/gt/5.wav" autoplay="">Your browser does not support the audio element.</audio></td>
1900                      <td style="text-align: center"><audio controls="controls" style="width: 160px;"><source src="audios/librispeech_16kHz/megatts/5.wav" autoplay="">Your browser does not support the audio element.</audio></td>
1901                      <td style="text-align: center"><audio controls="controls" style="width: 160px;"><source src="audios/librispeech_16kHz/ns2/5.wav" autoplay="">Your browser does not support the audio element.</audio></td>
1902                      <td style="text-align: center"><audio controls="controls" style="width: 160px;"><source src="audios/librispeech_16kHz/valle/5.wav" autoplay="">Your browser does not support the audio element.</audio></td>
1903                      <td style="text-align: center"><audio controls="controls" style="width: 160px;"><source src="audios/librispeech_16kHz/clam/5.wav" autoplay="">Your browser does not support the audio element.</audio></td>
1904                      <td style="text-align: center"><audio controls="controls" style="width: 160px;"><source src="audios/librispeech_16kHz/ours/5.wav" autoplay="">Your browser does not support the audio element.</audio></td>
1905                    </tr>
1906                    <tr>
1907                      <td style="text-align: left;vertical-align:middle;width: 500px">Thus did this humane and right minded father comfort his unhappy daughter, and her mother embracing her again, did all she could to soothe her feelings.</td>
1908                      <td style="text-align: center"><audio controls="controls" style="width: 160px;"><source src="audios/librispeech_16kHz/prompt/6.wav" autoplay="">Your browser does not support the audio element.</audio></td>
1909                      <td style="text-align: center"><audio controls="controls" style="width: 160px;"><source src="audios/librispeech_16kHz/gt/6.wav" autoplay="">Your browser does not support the audio element.</audio></td>
1910                      <td style="text-align: center; font-size: 18px; line-height: 2.5; padding-top: 10px; padding-bottom: 10px;">&mdash;</td>
1911                      <td style="text-align: center"><audio controls="controls" style="width: 160px;"><source src="audios/librispeech_16kHz/ns2/6.wav" autoplay="">Your browser does not support the audio element.</audio></td>
1912                      <td style="text-align: center"><audio controls="controls" style="width: 160px;"><source src="audios/librispeech_16kHz/valle/6.wav" autoplay="">Your browser does not support the audio element.</audio></td>
1913                      <td style="text-align: center"><audio controls="controls" style="width: 160px;"><source src="audios/librispeech_16kHz/clam/6.wav" autoplay="">Your browser does not support the audio element.</audio></td>
1914                      <td style="text-align: center"><audio controls="controls" style="width: 160px;"><source src="audios/librispeech_16kHz/ours/6.wav" autoplay="">Your browser does not support the audio element.</audio></td>
1915                    </tr>
1916                    <tr>
1917                      <td style="text-align: left;vertical-align:middle;width: 500px">He was in deep converse with the clerk and entered the hall holding him by the arm.</td>
1918                      <td style="text-align: center"><audio controls="controls" style="width: 160px;"><source src="audios/librispeech_16kHz/prompt/7.wav" autoplay="">Your browser does not support the audio element.</audio></td>
1919                      <td style="text-align: center"><audio controls="controls" style="width: 160px;"><source src="audios/librispeech_16kHz/gt/7.wav" autoplay="">Your browser does not support the audio element.</audio></td>
1920                      <td style="text-align: center"><audio controls="controls" style="width: 160px;"><source src="audios/librispeech_16kHz/megatts/7.wav" autoplay="">Your browser does not support the audio element.</audio></td>
1921                      <td style="text-align: center; font-size: 18px; line-height: 2.5; padding-top: 10px; padding-bottom: 10px;">&mdash;</td>
1922                      <td style="text-align: center"><audio controls="controls" style="width: 160px;"><source src="audios/librispeech_16kHz/valle/7.wav" autoplay="">Your browser does not support the audio element.</audio></td>
1923                      <td style="text-align: center"><audio controls="controls" style="width: 160px;"><source src="audios/librispeech_16kHz/clam/7.wav" autoplay="">Your browser does not support the audio element.</audio></td>
1924                      <td style="text-align: center"><audio controls="controls" style="width: 160px;"><source src="audios/librispeech_16kHz/ours/7.wav" autoplay="">Your browser does not support the audio element.</audio></td>
1925                    </tr>
1926                    <tr>
1927                      <td style="text-align: left;vertical-align:middle;width: 500px">
1927Indeed, there were only one or two strangers who could be admitted among the sisters without producing the same result.</td>
1928                      <td style="text-align: center"><audio controls="controls" style="width: 160px;"><source src="audios/librispeech_16kHz/prompt/8.wav" autoplay="">Your browser does not support the audio element.</audio></td>
1929                      <td style="text-align: center"><audio controls="controls" style="width: 160px;"><source src="audios/librispeech_16kHz/gt/8.wav" autoplay="">Your browser does not support the audio element.</audio></td>
1930                      <td style="text-align: center; font-size: 18px; line-height: 2.5; padding-top: 10px; padding-bottom: 10px;">&mdash;</td>
1931                      <td style="text-align: center"><audio controls="controls" style="width: 160px;"><source src="audios/librispeech_16kHz/ns2/8.wav" autoplay="">Your browser does not support the audio element.</audio></td>
1932                      <td style="text-align: center; font-size: 18px; line-height: 2.5; padding-top: 10px; padding-bottom: 10px;">&mdash;</td>
1933                      <td style="text-align: center"><audio controls="controls" style="width: 160px;"><source src="audios/librispeech_16kHz/clam/8.wav" autoplay="">Your browser does not support the audio element.</audio></td>
1934                      <td style="text-align: center"><audio controls="controls" style="width: 160px;"><source src="audios/librispeech_16kHz/ours/8.wav" autoplay="">Your browser does not support the audio element.</audio></td>
1935                    </tr>
1936                    <tr>
1937                      <td style="text-align: left;vertical-align:middle;width: 500px">For if he's anywhere on the farm, we can send for him in a minute.</td>
1938                      <td style="text-align: center"><audio controls="controls" style="width: 160px;"><source src="audios/librispeech_16kHz/prompt/9.wav" autoplay="">Your browser does not support the audio element.</audio></td>
1939                      <td style="text-align: center"><audio controls="controls" style="width: 160px;"><source src="audios/librispeech_16kHz/gt/9.wav" autoplay="">Your browser does not support the audio element.</audio></td>
1940                      <td style="text-align: center; font-size: 18px; line-height: 2.5; padding-top: 10px; padding-bottom: 10px;">&mdash;</td>
1941                      <td style="text-align: center"><audio controls="controls" style="width: 160px;"><source src="audios/librispeech_16kHz/ns2/9.wav" autoplay="">Your browser does not support the audio element.</audio></td>
1942                      <td style="text-align: center; font-size: 18px; line-height: 2.5; padding-top: 10px; padding-bottom: 10px;">&mdash;</td>
1943                      <td style="text-align: center"><audio controls="controls" style="width: 160px;"><source src="audios/librispeech_16kHz/clam/9.wav" autoplay="">Your browser does not support the audio element.</audio></td>
1944                      <td style="text-align: center"><audio controls="controls" style="width: 160px;"><source src="audios/librispeech_16kHz/ours/9.wav" autoplay="">Your browser does not support the audio element.</audio></td>
1945                    </tr>
1946                    <tr>
1947                      <td style="text-align: left;vertical-align:middle;width: 500px">Their piety would be like their names, like their faces, like their clothes, and it was idle for him to tell himself that their humble and contrite hearts it might be paid a far-richer tribute of devotion than his had ever been. A gift tenfold more acceptable than his elaborate adoration.</td>
1948                      <td style="text-align: center"><audio controls="controls" style="width: 160px;"><source src="audios/librispeech_16kHz/prompt/10.wav" autoplay="">Your browser does not support the audio element.</audio></td>
1949                      <td style="text-align: center"><audio controls="controls" style="width: 160px;"><source src="audios/librispeech_16kHz/gt/10.wav" autoplay="">Your browser does not support the audio element.</audio></td>
1950                      <td style="text-align: center; font-size: 18px; line-height: 2.5; padding-top: 10px; padding-bottom: 10px;">&mdash;</td>
1951                      <td style="text-align: center"><audio controls="controls" style="width: 160px;"><source src="audios/librispeech_16kHz/ns2/10.wav" autoplay="">Your browser does not support the audio element.</audio></td>
1952                      <td style="text-align: center; font-size: 18px; line-height: 2.5; padding-top: 10px; padding-bottom: 10px;">&mdash;</td>
1953                      <td style="text-align: center"><audio controls="controls" style="width: 160px;"><source src="audios/librispeech_16kHz/clam/10.wav" autoplay="">Your browser does not support the audio element.</audio></td>
1954                      <td style="text-align: center"><audio controls="controls" style="width: 160px;"><source src="audios/librispeech_16kHz/ours/10.wav" autoplay="">Your browser does not support the audio element.</audio></td>
1955                    </tr>
1956                    <tr>
1957                      <td style="text-align: left;vertical-align:middle;width: 500px">The air and the earth are curiously mated and intermingled as if the one were the breath of the other.</td>
1958                      <td style="text-align: center"><audio controls="controls" style="width: 160px;"><source src="audios/librispeech_16kHz/prompt/11.wav" autoplay="">Your browser does not support the audio element.</audio></td>
1959                      <td style="text-align: center"><audio controls="controls" style="width: 160px;"><source src="audios/librispeech_16kHz/gt/11.wav" autoplay="">Your browser does not support the audio element.</audio></td>
1960                      <td style="text-align: center; font-size: 18px; line-height: 2.5; padding-top: 10px; padding-bottom: 10px;">&mdash;</td>
1961                      <td style="text-align: center"><audio controls="controls" style="width: 160px;"><source src="audios/librispeech_16kHz/ns2/11.wav" autoplay="">Your browser does not support the audio element.</audio></td>
1962                      <td style="text-align: center; font-size: 18px; line-height: 2.5; padding-top: 10px; padding-bottom: 10px;">&mdash;</td>
1963                      <td style="text-align: center"><audio controls="controls" style="width: 160px;"><source src="audios/librispeech_16kHz/clam/11.wav" autoplay="">Your browser does not support the audio element.</audio></td>
1964                      <td style="text-align: center"><audio controls="controls" style="width: 160px;"><source src="audios/librispeech_16kHz/ours/11.wav" autoplay="">Your browser does not support the audio element.</audio></td>
1965                    </tr>
1966                    <tr>
1967                      <td style="text-align: left;vertical-align:middle;width: 500px">I had always known him to be restless in his manner, but on this particular occasion he was in such a state of uncontrollable agitation that it was clear something very unusual had occurred.</td>
1968                      <td style="text-align: center"><audio controls="controls" style="width: 160px;"><source src="audios/librispeech_16kHz/prompt/12.wav" autoplay="">Your browser does not support the audio element.</audio></td>
1969                      <td style="text-align: center"><audio controls="controls" style="width: 160px;"><source src="audios/librispeech_16kHz/gt/12.wav" autoplay="">Your browser does not support the audio element.</audio></td>
1970                      <td style="text-align: center; font-size: 18px; line-height: 2.5; padding-top: 10px; padding-bottom: 10px;">&mdash;</td>
1971                      <td style="text-align: center"><audio controls="controls" style="width: 160px;"><source src="audios/librispeech_16kHz/ns2/12.wav" autoplay="">Your browser does not support the audio element.</audio></td>
1972                      <td style="text-align: center; font-size: 18px; line-height: 2.5; padding-top: 10px; padding-bottom: 10px;">&mdash;</td>
1973                      <td style="text-align: center"><audio controls="controls" style="width: 160px;"><source src="audios/librispeech_16kHz/clam/12.wav" autoplay="">Your browser does not support the audio element.</audio></td>
1974                      <td style="text-align: center"><audio controls="controls" style="width: 160px;"><source src="audios/librispeech_16kHz/ours/12.wav" autoplay="">Your browser does not support the audio element.</audio></td>
1975                    </tr>
1976                    <tr>
1977                      <td style="text-align: left;vertical-align:middle;width: 500px">His death in this conjuncture was a public misfortune.</td>
1978                      <td style="text-align: center"><audio controls="controls" style="width: 160px;">
1978<source src="audios/librispeech_16kHz/prompt/13.wav" autoplay="">Your browser does not support the audio element.</audio></td>
1979                      <td style="text-align: center"><audio controls="controls" style="width: 160px;"><source src="audios/librispeech_16kHz/gt/13.wav" autoplay="">Your browser does not support the audio element.</audio></td>
1980                      <td style="text-align: center; font-size: 18px; line-height: 2.5; padding-top: 10px; padding-bottom: 10px;">&mdash;</td>
1981                      <td style="text-align: center"><audio controls="controls" style="width: 160px;"><source src="audios/librispeech_16kHz/ns2/13.wav" autoplay="">Your browser does not support the audio element.</audio></td>
1982                      <td style="text-align: center; font-size: 18px; line-height: 2.5; padding-top: 10px; padding-bottom: 10px;">&mdash;</td>
1983                      <td style="text-align: center"><audio controls="controls" style="width: 160px;"><source src="audios/librispeech_16kHz/clam/13.wav" autoplay="">Your browser does not support the audio element.</audio></td>
1984                      <td style="text-align: center"><audio controls="controls" style="width: 160px;"><source src="audios/librispeech_16kHz/ours/13.wav" autoplay="">Your browser does not support the audio element.</audio></td>
1985                    </tr>
1986                    <tr>
1987                      <td style="text-align: left;vertical-align:middle;width: 500px">It is this that is of interest to theory of knowledge.</td>
1988                      <td style="text-align: center"><audio controls="controls" style="width: 160px;"><source src="audios/librispeech_16kHz/prompt/14.wav" autoplay="">Your browser does not support the audio element.</audio></td>
1989                      <td style="text-align: center"><audio controls="controls" style="width: 160px;"><source src="audios/librispeech_16kHz/gt/14.wav" autoplay="">Your browser does not support the audio element.</audio></td>
1990                      <td style="text-align: center; font-size: 18px; line-height: 2.5; padding-top: 10px; padding-bottom: 10px;">&mdash;</td>
1991                      <td style="text-align: center"><audio controls="controls" style="width: 160px;"><source src="audios/librispeech_16kHz/ns2/14.wav" autoplay="">Your browser does not support the audio element.</audio></td>
1992                      <td style="text-align: center; font-size: 18px; line-height: 2.5; padding-top: 10px; padding-bottom: 10px;">&mdash;</td>
1993                      <td style="text-align: center"><audio controls="controls" style="width: 160px;"><source src="audios/librispeech_16kHz/clam/14.wav" autoplay="">Your browser does not support the audio element.</audio></td>
1994                      <td style="text-align: center"><audio controls="controls" style="width: 160px;"><source src="audios/librispeech_16kHz/ours/14.wav" autoplay="">Your browser does not support the audio element.</audio></td>
1995                    </tr>
1996                    <tr>
1997                      <td style="text-align: left;vertical-align:middle;width: 500px">For a few miles, she followed the line hitherto presumably occupied by the coast of Algeria, but no land appeared to the south.</td>
1998                      <td style="text-align: center"><audio controls="controls" style="width: 160px;"><source src="audios/librispeech_16kHz/prompt/15.wav" autoplay="">Your browser does not support the audio element.</audio></td>
1999                      <td style="text-align: center"><audio controls="controls" style="width: 160px;"><source src="audios/librispeech_16kHz/gt/15.wav" autoplay="">Your browser does not support the audio element.</audio></td>
2000                      <td style="text-align: center; font-size: 18px; line-height: 2.5; padding-top: 10px; padding-bottom: 10px;">&mdash;</td>
2001                      <td style="text-align: center"><audio controls="controls" style="width: 160px;"><source src="audios/librispeech_16kHz/ns2/15.wav" autoplay="">Your browser does not support the audio element.</audio></td>
2002                      <td style="text-align: center; font-size: 18px; line-height: 2.5; padding-top: 10px; padding-bottom: 10px;">&mdash;</td>
2003                      <td style="text-align: center"><audio controls="controls" style="width: 160px;"><source src="audios/librispeech_16kHz/clam/15.wav" autoplay="">Your browser does not support the audio element.</audio></td>
2004                      <td style="text-align: center"><audio controls="controls" style="width: 160px;"><source src="audios/librispeech_16kHz/ours/15.wav" autoplay="">Your browser does not support the audio element.</audio></td>
2005                    </tr>
2006                    </tbody>
2007                    </table>
2008                  </div>
2009              </div>
2010
2011            </div>
2012          </div>
2013        </article>
2014      </main>
2015    </div>
2016    <template id="prompt-template">
2017      <div class="card card-body ditto-card">
2018        <div>
2019          <p class="ditto-prompt-text">
2020            <strong>Prompt Text</strong>: <span></span>
2021          </p>
2022          <p class="ditto-text"><strong>Text</strong>: <span></span></p>
2023        </div>
2024        <div class="ditto-audioviz">
2025          <canvas></canvas>
2026          <audio controls=""></audio>
2027        </div>
2028        <div class="ditto-toggle">
2029          <div class="show">
2030            <i class="bi bi-arrow-down-square"></i>
2031            Show baseline(s)
2032          </div>
2033          <div>
2034            <i class="bi bi-arrow-up-square" style="margin-right: 6.34px"></i>
2035            Hide baseline(s)
2036          </div>
2037        </div>
2038        <div class="collapse multi-collapse ditto-sample-box">
2039          <div class="table-responsive pt-3">
2040            <table class="table">
2041              <thead>
2042                <tr>
2043                  <th class="border-right">Prompt (Unmodified)</th>
2044                  <th>Ground Truth</th>
2045                  <th>YourTTS</th>
2046                  <th>Vall-E</th>
2047                  <th>CLaM-TTS</th>
2048                  <th>Mega-TTS</th>
2049                </tr>
2050              </thead>
2051              <tbody>
2052                <tr>
2053                  <td class="border-right" data-label="Prompt (Unmodified)">
2054                    <audio controls="controls" preload="none"></audio>
2055                  </td>
2056                  <td data-label="Ground Truth">
2057                    <audio controls="controls" preload="none"></audio>
2058                  </td>
2059                  <td data-label="YourTTS">
2060                    <audio controls="controls" preload="none"></audio>
2061                  </td>
2062                  <td data-label="Vall-E">
2063                    <audio controls="controls" preload="none"></audio>
2064                  </td>
2065                  <td data-label="CLaM-TTS">
2066                    <audio controls="controls" preload="none"></audio>
2067                  </td>
2068                  <td data-label="Mega-TTS">
2069                    <audio controls="controls" preload="none"></audio>
2070                  </td>
2071                </tr>
2072              </tbody>
2073            </table>
2074          </div>
2075        </div>
2076      </div>
2077    </template>
2078    <div id="fab" class="fab">
2079      <a class="mainop" href="#myTab">
2080        <i class="bi bi-arrow-bar-up"></i>
2081      </a>
2082    </div>
2083  </body>
2084</html>

Line numbers count LF bytes from the start of the resource, as the search results do. Vendor segments are library code the classifier recognised; they are stored but not indexed. Bytes are shown as Latin1 characters, one per byte.