ming
/
TTS_fs2


			
				
					
						
						
							123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960616263646566676869707172737475767778798081828384858687888990919293949596979899100101102103104105106107108109110111112113114115116117118119120121122123124125126127128129130131132133134135136137138139140141142143144145146147148149150151152153154155156157158159160161162163164165166167168169170171172173174175176177178179180181182183184185186187188189190191192193194195196197198199200201202203204205206207208209210211212213214215216217218219220221222223224225226227228229230231232233234235236237238239240241242243244245246247248249250251252253254255256257258259260261262263264265266267268269270271272273274275276277278279280281282283284285286287288289290291292293294295296297298299300301302303304305306307308309310311312313314315316317318319320321322323324325326327328329330331332333334335336337338339340341342343344345346347348349350351352353354355356357358359360361362363364365366367368369370371372373374375376377378379380381382383384385386387388389390391392393394395396397398399400401402403404405406407408409410411412413414415416417418419420421422423424425426427428429430431432433434435436437438439440441442443444445446447448449450451452453454455456457458459460461462463464465466467468469470471472473474475476477478479480481482483484485486487488489490491492493494495496497498499500501502503504505506507508509510511512513514515516517518519520521522523524525526527528529530531532533534535536537538539540541542543544545546547548549550551552553554555556557558559560561562563564565566567568569570571572573574575576577578579580581582583584585586587588589590591592593594595596597598599600601602603604605606607608609610611612613614615616617618619620621622623624625626627628629630631632633634635636637638639640641642643644645646647648649650651652653654655656657658659660661662663664665666667668669670671672673674675676677678679680681682683684685686687688689690691692693694695696697698699700701702703704705706707708709710711712713714715716717718719720721
							<!DOCTYPE html>
<html>

<head>
  <meta charset="utf-8" />
  <meta name="viewport" content="width=device-width,initial-scale=1" />
  <title>FastSpeech 2 Audio Samples</title>
  <style>
    body {
      margin: 0% 15%;
      padding: 50px 30px;
      background: #fff;
      color: #111;
      font-size: 17px;
      font-family: sans-serif;
      font-weight: 400;
      line-height: 1.8;
      overflow-x: hidden;
      -webkit-font-smoothing: antialiased;
    }

    h1 {
      font-size: 1.75em;
    }

    h2 {
      margin-bottom: 0.4em;
    }

    hr {
      height: 0.5px;
      border-width: 0;
      color: lightgray;
      background-color: lightgray
    }

    table {
      width: 100%;
    }

    audio {
      width: 100%;
    }
  </style>
</head>

<body>
  <h1>FastSpeech 2 Audio Samples</h1>
  <p>
  <h2>English Single-Speaker TTS</h2>
  <div>
    <b>Dataset</b>: <a href="https://keithito.com/LJ-Speech-Dataset/">LJSpeech</a>
  </div>
  <div>
    <b>Checkpoint</b>: <a
      href="https://drive.google.com/file/d/1r3fYhnblBJ8hDKDSUDtidJ-BN-xAM9pe/view?usp=sharing">link</a>
  </div>
  <div>
    <b>Config</b>: <a href="https://github.com/ming024/FastSpeech2/tree/master/config/LJSpeech">link</a>
  </div>
  <div>
    <b>Vocoder</b>: <a href="https://github.com/jik876/hifi-gan">HiFi-GAN (LJSpeech)</a>
  </div>

  <hr>

  <table>
    <tr>
      <th style="text-align: center">Extremely Long Sentence</th>
    </tr>
    <tr>
      <td style="text-align: center"><audio controls="controls">
          <source src="./demo/LJSpeech/LJSpeech_long_sentence.wav" autoplay />
        </audio></td>
    </tr>
  </table>
  <div><b>Text</b>: <em>Advanced text-to-speech models such as FastSpeech can synthesize speech significantly faster
      than previous autoregressive models with comparable quality. The training of FastSpeech model relies on an
      autoregressive teacher model for duration prediction and knowledge distillation , which can ease the one-to-many
      mapping problem in TTS. However, FastSpeech has several disadvantages: 1) the teacher-student distillation
      pipeline is complicated, 2) the duration extracted from the teacher model is not accurate enough, and the target
      mel-spectrograms distilled from teacher model suffer from information loss due to data simplification, both of
      which limit the voice quality. In this paper, we propose FastSpeech 2, which addresses the issues in FastSpeech
      and better solves the one-to-many mapping problem in TTS by 1) directly training the model with ground-truth
      target instead of the simplified output from teacher, and 2) introducing more variation information of speech as
      conditional inputs. Specifically, we extract duration, pitch and energy from speech waveform and directly take
      them as conditional inputs during training and use predicted values during inference. We further design FastSpeech
      2s, which is the first attempt to directly generate speech waveform from text in parallel, enjoying the benefit of
      full end-to-end training and even faster inference than FastSpeech. Experimental results show that 1) FastSpeech 2
      and 2s outperform FastSpeech in voice quality with much simplified training pipeline and reduced training time; 2)
      FastSpeech 2 and 2s can match the voice quality of autoregressive models while enjoying much faster inference
      speed.
    </em></div>

  <hr>

  <table>
    <tr>
      <th style="text-align: center">LJ042-0094 (Ground-Truth) </th>
      <th style="text-align: center">LJ042-0094 (Synthesized) </th>
    </tr>
    <tr>
      <td style="text-align: center"><audio controls="controls">
          <source src="./demo/LJSpeech/LJ042-0094_ground-truth.wav" autoplay />
        </audio></td>
      <td style="text-align: center"><audio controls="controls">
          <source src="./demo/LJSpeech/LJ042-0094_synthesized.wav" autoplay />
        </audio></td>
    </tr>
  </table>
  <div><b>Text</b>: <em>the soviet authorities denied oswald permission</em></div>

  <hr>

  <table>
    <tr>
      <th style="text-align: center">LJ021-0108 (Ground-Truth) </th>
      <th style="text-align: center">LJ021-0108 (Synthesized) </th>
    </tr>
    <tr>
      <td style="text-align: center"><audio controls="controls">
          <source src="./demo/LJSpeech/LJ021-0108_ground-truth.wav" autoplay />
        </audio></td>
      <td style="text-align: center"><audio controls="controls">
          <source src="./demo/LJSpeech/LJ021-0108_synthesized.wav" autoplay />
        </audio></td>
    </tr>
  </table>
  <div><b>Text</b>: <em>from the standpoint of the good of the industries themselves, as well as the general public
      interest,</em></div>

  <hr>

  <table>
    <tr>
      <th style="text-align: center">LJ030-0031 (Ground-Truth) </th>
      <th style="text-align: center">LJ030-0031 (Synthesized) </th>
    </tr>
    <tr>
      <td style="text-align: center"><audio controls="controls">
          <source src="./demo/LJSpeech/LJ030-0031_ground-truth.wav" autoplay />
        </audio></td>
      <td style="text-align: center"><audio controls="controls">
          <source src="./demo/LJSpeech/LJ030-0031_synthesized.wav" autoplay />
        </audio></td>
    </tr>
  </table>
  <div><b>Text</b>: <em>secret service agents formed a cordon to keep the press and photographers from impeding their
      passage and scanned the
      crowd for threatening movements.</em></div>

  <hr>

  <table>
    <tr>
      <th style="text-align: center">LJ001-0012 (Ground-Truth) </th>
      <th style="text-align: center">LJ001-0012 (Synthesized) </th>
    </tr>
    <tr>
      <td style="text-align: center"><audio controls="controls">
          <source src="./demo/LJSpeech/LJ001-0012_ground-truth.wav" autoplay />
        </audio></td>
      <td style="text-align: center"><audio controls="controls">
          <source src="./demo/LJSpeech/LJ001-0012_synthesized.wav" autoplay />
        </audio></td>
    </tr>
  </table>
  <div><b>Text</b>: <em>especially as no more time is occupied, or cost incurred, in casting, setting, or printing
      beautiful letters</em></div>

  <hr>

  <table>
    <tr>
      <th style="text-align: center">LJ011-0202 (Ground-Truth) </th>
      <th style="text-align: center">LJ011-0202 (Synthesized) </th>
    </tr>
    <tr>
      <td style="text-align: center"><audio controls="controls">
          <source src="./demo/LJSpeech/LJ011-0202_ground-truth.wav" autoplay />
        </audio></td>
      <td style="text-align: center"><audio controls="controls">
          <source src="./demo/LJSpeech/LJ011-0202_synthesized.wav" autoplay />
        </audio></td>
    </tr>
  </table>
  <div><b>Text</b>: <em>the uncle claimed her. the husband resisted.</em></div>

  <hr>

  <table>
    <tr>
      <th style="text-align: center">LJ006-0114 (Ground-Truth) </th>
      <th style="text-align: center">LJ006-0114 (Synthesized) </th>
    </tr>
    <tr>
      <td style="text-align: center"><audio controls="controls">
          <source src="./demo/LJSpeech/LJ006-0114_ground-truth.wav" autoplay />
        </audio></td>
      <td style="text-align: center"><audio controls="controls">
          <source src="./demo/LJSpeech/LJ006-0114_synthesized.wav" autoplay />
        </audio></td>
    </tr>
  </table>
  <div><b>Text</b>: <em>they bought their offices from one another, and were thus considered to have a vested interest
      in them.</em></div>

  <hr>

  <table>
    <tr>
      <th style="text-align: center">LJ009-0038 (Ground-Truth) </th>
      <th style="text-align: center">LJ009-0038 (Synthesized) </th>
    </tr>
    <tr>
      <td style="text-align: center"><audio controls="controls">
          <source src="./demo/LJSpeech/LJ009-0038_ground-truth.wav" autoplay />
        </audio></td>
      <td style="text-align: center"><audio controls="controls">
          <source src="./demo/LJSpeech/LJ009-0038_synthesized.wav" autoplay />
        </audio></td>
    </tr>
  </table>
  <div><b>Text</b>: <em>in the center of the chapel was the condemned pew, a large dock-like erection painted
      black.</em></div>

  <hr>

  <table>
    <tr>
      <th style="text-align: center">LJ006-0161 (Ground-Truth) </th>
      <th style="text-align: center">LJ006-0161 (Synthesized) </th>
    </tr>
    <tr>
      <td style="text-align: center"><audio controls="controls">
          <source src="./demo/LJSpeech/LJ006-0161_ground-truth.wav" autoplay />
        </audio></td>
      <td style="text-align: center"><audio controls="controls">
          <source src="./demo/LJSpeech/LJ006-0161_synthesized.wav" autoplay />
        </audio></td>
    </tr>
  </table>
  <div><b>Text</b>: <em>again, a turnkey deposed that his chief did not enter the wards more than once a fortnight.</em>
  </div>

  <hr>

  <table>
    <tr>
      <th style="text-align: center">LJ027-0127 (Ground-Truth) </th>
      <th style="text-align: center">LJ027-0127 (Synthesized) </th>
    </tr>
    <tr>
      <td style="text-align: center"><audio controls="controls">
          <source src="./demo/LJSpeech/LJ027-0127_ground-truth.wav" autoplay />
        </audio></td>
      <td style="text-align: center"><audio controls="controls">
          <source src="./demo/LJSpeech/LJ027-0127_synthesized.wav" autoplay />
        </audio></td>
    </tr>
  </table>
  <div><b>Text</b>: <em>while neglecting to maintain his unity of ideal in the case of nearly all the numerous species
      of snakes, he should have
      added a tiny rudiment in the case of the python</em></div>

  <hr>

  <table>
    <tr>
      <th style="text-align: center">LJ050-0158 (Ground-Truth) </th>
      <th style="text-align: center">LJ050-0158 (Synthesized) </th>
    </tr>
    <tr>
      <td style="text-align: center"><audio controls="controls">
          <source src="./demo/LJSpeech/LJ050-0158_ground-truth.wav" autoplay />
        </audio></td>
      <td style="text-align: center"><audio controls="controls">
          <source src="./demo/LJSpeech/LJ050-0158_synthesized.wav" autoplay />
        </audio></td>
    </tr>
  </table>
  <div><b>Text</b>: <em>the department hopes to design a practical system which will fully meet the needs of the
      protective research section of
      the secret service.</em></div>

  </p>

  <hr>

  <p>
  <h2>Mandarin Multi-Speaker TTS</h2>
  <div>
    <b>Dataset</b>: <a href="http://www.aishelltech.com/aishell_3">AISHELL-3</a>
  </div>
  <div>
    <b>Checkpoint</b>: <a
      href="https://drive.google.com/file/d/1uYWd5JlaK-fochQ2JFgIP_wOEkoLPLqs/view?usp=sharing">link</a>
  </div>
  <div>
    <b>Config</b>: <a href="https://github.com/ming024/FastSpeech2/tree/master/config/AISHELL3">link</a>
  </div>
  <div>
    <b>Vocoder</b>: <a href="https://github.com/jik876/hifi-gan">HiFi-GAN (universal)</a>
  </div>

  <hr>

  <table>
    <tr>
      <th style="text-align: center">SSB08630110 (Ground-Truth) </th>
      <th style="text-align: center">SSB08630110 (Synthesized) </th>
    </tr>
    <tr>
      <td style="text-align: center"><audio controls="controls">
          <source src="./demo/AISHELL3/SSB08630110_ground-truth.wav" autoplay />
        </audio></td>
      <td style="text-align: center"><audio controls="controls">
          <source src="./demo/AISHELL3/SSB08630110_synthesized.wav" autoplay />
        </audio></td>
    </tr>
  </table>
  <div><b>Speaker</b>: <em>SSB0863</em></div>
  <div><b>Text</b>: <em>放首歌来听，丁当的歌 (fang4 shou3 ge1 lai2 ting1 sp ding1 dang1 de5 ge1)</em></div>

  <hr>

  <table>
    <tr>
      <th style="text-align: center">SSB07940279 (Ground-Truth) </th>
      <th style="text-align: center">SSB07940279 (Synthesized) </th>
    </tr>
    <tr>
      <td style="text-align: center"><audio controls="controls">
          <source src="./demo/AISHELL3/SSB07940279_ground-truth.wav" autoplay />
        </audio></td>
      <td style="text-align: center"><audio controls="controls">
          <source src="./demo/AISHELL3/SSB07940279_synthesized.wav" autoplay />
        </audio></td>
    </tr>
  </table>
  <div><b>Speaker</b>: <em>SSB0794</em></div>
  <div><b>Text</b>: <em>中国，今日有望创下历史单，日，出行人次最高 (zhong1 guo2 sp jin1 ri4 you3 wang4 chuang4 xia4 li4 shi3 dan1 sp ri4 sp
      chu1 xing2 ren2 ci4 zui4 gao1)</em></div>

  <hr>

  <table>
    <tr>
      <th style="text-align: center">SSB00800003 (Ground-Truth) </th>
      <th style="text-align: center">SSB00800003 (Synthesized) </th>
    </tr>
    <tr>
      <td style="text-align: center"><audio controls="controls">
          <source src="./demo/AISHELL3/SSB00800003_ground-truth.wav" autoplay />
        </audio></td>
      <td style="text-align: center"><audio controls="controls">
          <source src="./demo/AISHELL3/SSB00800003_synthesized.wav" autoplay />
        </audio></td>
    </tr>
  </table>
  <div><b>Speaker</b>: <em>SSSB0080</em></div>
  <div><b>Text</b>: <em>中新网七月十四日电 (zhong1 xin1 wang3 qi1 yue4 shi2 si4 ri4 dian4)</em></div>

  <hr>

  <table>
    <tr>
      <th style="text-align: center">SSB15930399 (Ground-Truth) </th>
      <th style="text-align: center">SSB15930399 (Synthesized) </th>
    </tr>
    <tr>
      <td style="text-align: center"><audio controls="controls">
          <source src="./demo/AISHELL3/SSB15930399_ground-truth.wav" autoplay />
        </audio></td>
      <td style="text-align: center"><audio controls="controls">
          <source src="./demo/AISHELL3/SSB15930399_synthesized.wav" autoplay />
        </audio></td>
    </tr>
  </table>
  <div><b>Speaker</b>: <em>SSB1593</em></div>
  <div><b>Text</b>: <em>也就是说汤米踩过点 (ye3 jiu4 shi4 shuo1 tang1 mi3 cai3 guo4 dian3)</em></div>

  <hr>

  <table>
    <tr>
      <th style="text-align: center">SSB16070080 (Ground-Truth) </th>
      <th style="text-align: center">SSB16070080 (Synthesized) </th>
    </tr>
    <tr>
      <td style="text-align: center"><audio controls="controls">
          <source src="./demo/AISHELL3/SSB16070080_ground-truth.wav" autoplay />
        </audio></td>
      <td style="text-align: center"><audio controls="controls">
          <source src="./demo/AISHELL3/SSB16070080_synthesized.wav" autoplay />
        </audio></td>
    </tr>
  </table>
  <div><b>Speaker</b>: <em>SSB1607</em></div>
  <div><b>Text</b>: <em>但阵营之间差距拉大 (dan4 zhen4 ying2 zhi1 jian1 cha1 ju4 la1 da4)</em></div>

  <hr>

  <table>
    <tr>
      <th style="text-align: center">SSB10200388 (Ground-Truth) </th>
      <th style="text-align: center">SSB10200388 (Synthesized) </th>
    </tr>
    <tr>
      <td style="text-align: center"><audio controls="controls">
          <source src="./demo/AISHELL3/SSB10200388_ground-truth.wav" autoplay />
        </audio></td>
      <td style="text-align: center"><audio controls="controls">
          <source src="./demo/AISHELL3/SSB10200388_synthesized.wav" autoplay />
        </audio></td>
    </tr>
  </table>
  <div><b>Speaker</b>: <em>SSSB1020</em></div>
  <div><b>Text</b>: <em>我抱着一大堆纸箱回了家 (wo3 bao4 zhe5 yi2 da4 dui1 zhi3 xiang1 hui2 le5 jia1)</em></div>

  <hr>

  <table>
    <tr>
      <th style="text-align: center">SSB05900363 (Ground-Truth) </th>
      <th style="text-align: center">SSB05900363 (Synthesized) </th>
    </tr>
    <tr>
      <td style="text-align: center"><audio controls="controls">
          <source src="./demo/AISHELL3/SSB05900363_ground-truth.wav" autoplay />
        </audio></td>
      <td style="text-align: center"><audio controls="controls">
          <source src="./demo/AISHELL3/SSB05900363_synthesized.wav" autoplay />
        </audio></td>
    </tr>
  </table>
  <div><b>Speaker</b>: <em>SSB0590</em></div>
  <div><b>Text</b>: <em>我们不能分头行动 (wo3 men2 bu4 neng2 fen1 tou2 xing2 dong4)</em></div>

  <hr>

  <table>
    <tr>
      <th style="text-align: center">SSB04700028 (Ground-Truth) </th>
      <th style="text-align: center">SSB04700028 (Synthesized) </th>
    </tr>
    <tr>
      <td style="text-align: center"><audio controls="controls">
          <source src="./demo/AISHELL3/SSB04700028_ground-truth.wav" autoplay />
        </audio></td>
      <td style="text-align: center"><audio controls="controls">
          <source src="./demo/AISHELL3/SSB04700028_synthesized.wav" autoplay />
        </audio></td>
    </tr>
  </table>
  <div><b>Speaker</b>: <em>SSB0470</em></div>
  <div><b>Text</b>: <em>毕节市的景点有什么 (bi4 jie2 shi4 de5 jing2 dian2 you3 shen2 me5)</em></div>

  <hr>

  <table>
    <tr>
      <th style="text-align: center">SSB06290079 (Ground-Truth) </th>
      <th style="text-align: center">SSB06290079 (Synthesized) </th>
    </tr>
    <tr>
      <td style="text-align: center"><audio controls="controls">
          <source src="./demo/AISHELL3/SSB06290079_ground-truth.wav" autoplay />
        </audio></td>
      <td style="text-align: center"><audio controls="controls">
          <source src="./demo/AISHELL3/SSB06290079_synthesized.wav" autoplay />
        </audio></td>
    </tr>
  </table>
  <div><b>Speaker</b>: <em>SSB0629</em></div>
  <div><b>Text</b>: <em>你以为 (ni2 yi3 wei2)</em></div>

  <hr>

  <table>
    <tr>
      <th style="text-align: center">SSB08220199 (Ground-Truth) </th>
      <th style="text-align: center">SSB08220199 (Synthesized) </th>
    </tr>
    <tr>
      <td style="text-align: center"><audio controls="controls">
          <source src="./demo/AISHELL3/SSB08220199_ground-truth.wav" autoplay />
        </audio></td>
      <td style="text-align: center"><audio controls="controls">
          <source src="./demo/AISHELL3/SSB08220199_synthesized.wav" autoplay />
        </audio></td>
    </tr>
  </table>
  <div><b>Speaker</b>: <em>SSB0822</em></div>
  <div><b>Text</b>: <em>该职位员工可以，居住在西雅，图或，洛，杉矶 (gai1 zhi2 wei4 yuan2 gong1 ke2 yi3 sp ju1 zhu4 zai4 xi1 ya3 sp tu2 huo4
      sp luo4 sp shan3 ji1)</em></div>

  </p>

  <hr>

  <p>
  <h2>English Multi-Speaker TTS</h2>
  <div>
    <b>Dataset</b>: <a href="https://research.google/tools/datasets/libri-tts/">LibriTTS</a>
  </div>
  <div>
    <b>Checkpoint</b>: <a
      href="https://drive.google.com/file/d/1M6BxJtTxYW56dG1Myz9MqZmG_OXJLUqy/view?usp=sharing">link</a>
  </div>
  <div>
    <b>Config</b>: <a href="https://github.com/ming024/FastSpeech2/tree/master/config/LibriTTS">link</a>
  </div>
  <div>
    <b>Vocoder</b>: <a href="https://github.com/jik876/hifi-gan">HiFi-GAN (universal)</a>
  </div>

  <hr>

  <table>
    <tr>
      <th style="text-align: center">207_131203_000011_000000 (Ground-Truth) </th>
      <th style="text-align: center">207_131203_000011_000000 (Synthesized) </th>
    </tr>
    <tr>
      <td style="text-align: center"><audio controls="controls">
          <source src="./demo/LibriTTS/207_131203_000011_000000_ground-truth.wav" autoplay />
        </audio></td>
      <td style="text-align: center"><audio controls="controls">
          <source src="./demo/LibriTTS/207_131203_000011_000000_synthesized.wav" autoplay />
        </audio></td>
    </tr>
  </table>
  <div><b>Speaker</b>: <em>207</em></div>
  <div><b>Text</b>: <em>his mother, however, was a little shy of the company for him, and besides she could not always
      spare him.</em></div>

  <hr>

  <table>
    <tr>
      <th style="text-align: center">2299_6524_000057_000000 (Ground-Truth) </th>
      <th style="text-align: center">2299_6524_000057_000000 (Synthesized) </th>
    </tr>
    <tr>
      <td style="text-align: center"><audio controls="controls">
          <source src="./demo/LibriTTS/2299_6524_000057_000000_ground-truth.wav" autoplay />
        </audio></td>
      <td style="text-align: center"><audio controls="controls">
          <source src="./demo/LibriTTS/2299_6524_000057_000000_synthesized.wav" autoplay />
        </audio></td>
    </tr>
  </table>
  <div><b>Speaker</b>: <em>2299</em></div>
  <div><b>Text</b>: <em>on the arrival at the hut to my chagrin we found it filled with snow.</em></div>

  <hr>

  <table>
    <tr>
      <th style="text-align: center">2388_153731_000003_000000 (Ground-Truth) </th>
      <th style="text-align: center">2388_153731_000003_000000 (Synthesized) </th>
    </tr>
    <tr>
      <td style="text-align: center"><audio controls="controls">
          <source src="./demo/LibriTTS/2388_153731_000003_000000_ground-truth.wav" autoplay />
        </audio></td>
      <td style="text-align: center"><audio controls="controls">
          <source src="./demo/LibriTTS/2388_153731_000003_000000_synthesized.wav" autoplay />
        </audio></td>
    </tr>
  </table>
  <div><b>Speaker</b>: <em>2388</em></div>
  <div><b>Text</b>: <em>the story of the first scientific observation of the corona and the prominences is thrillingly
      interesting, and in fact dramatic.</em></div>

  <hr>

  <table>
    <tr>
      <th style="text-align: center">3615_14677_000014_000000 (Ground-Truth) </th>
      <th style="text-align: center">3615_14677_000014_000000 (Synthesized) </th>
    </tr>
    <tr>
      <td style="text-align: center"><audio controls="controls">
          <source src="./demo/LibriTTS/3615_14677_000014_000000_ground-truth.wav" autoplay />
        </audio></td>
      <td style="text-align: center"><audio controls="controls">
          <source src="./demo/LibriTTS/3615_14677_000014_000000_synthesized.wav" autoplay />
        </audio></td>
    </tr>
  </table>
  <div><b>Speaker</b>: <em>3615</em></div>
  <div><b>Text</b>: <em>beat three eggs with a pinch of salt; add one pint of milk and two thirds of a cup of
      flour.</em></div>

  <hr>

  <table>
    <tr>
      <th style="text-align: center">4744_31668_000003_000001 (Ground-Truth) </th>
      <th style="text-align: center">4744_31668_000003_000001 (Synthesized) </th>
    </tr>
    <tr>
      <td style="text-align: center"><audio controls="controls">
          <source src="./demo/LibriTTS/4744_31668_000003_000001_ground-truth.wav" autoplay />
        </audio></td>
      <td style="text-align: center"><audio controls="controls">
          <source src="./demo/LibriTTS/4744_31668_000003_000001_synthesized.wav" autoplay />
        </audio></td>
    </tr>
  </table>
  <div><b>Speaker</b>: <em>4744</em></div>
  <div><b>Text</b>: <em>far from morbid naturally, she did her best to deny the thought, and so simple and unartificial
      was her type of mind that for weeks together she would wholly lose it.</em></div>

  <hr>

  <table>
    <tr>
      <th style="text-align: center">480_127525_000006_000000 (Ground-Truth) </th>
      <th style="text-align: center">480_127525_000006_000000 (Synthesized) </th>
    </tr>
    <tr>
      <td style="text-align: center"><audio controls="controls">
          <source src="./demo/LibriTTS/480_127525_000006_000000_ground-truth.wav" autoplay />
        </audio></td>
      <td style="text-align: center"><audio controls="controls">
          <source src="./demo/LibriTTS/480_127525_000006_000000_synthesized.wav" autoplay />
        </audio></td>
    </tr>
  </table>
  <div><b>Speaker</b>: <em>480</em></div>
  <div><b>Text</b>: <em>i tried and found by experiment that the tide kept sweeping us westward until i had laid her
      head due east, or just about right angles to the way we ought to go.</em></div>

  <hr>

  <table>
    <tr>
      <th style="text-align: center">5126_27504_000011_000003 (Ground-Truth) </th>
      <th style="text-align: center">5126_27504_000011_000003 (Synthesized) </th>
    </tr>
    <tr>
      <td style="text-align: center"><audio controls="controls">
          <source src="./demo/LibriTTS/5126_27504_000011_000003_ground-truth.wav" autoplay />
        </audio></td>
      <td style="text-align: center"><audio controls="controls">
          <source src="./demo/LibriTTS/5126_27504_000011_000003_synthesized.wav" autoplay />
        </audio></td>
    </tr>
  </table>
  <div><b>Speaker</b>: <em>5126</em></div>
  <div><b>Text</b>: <em>it looked as if our luck was dead out, and we began to think our chance of getting across the
      border to queensland, and clear out of the colony that way, looked worse every day.</em></div>

  <hr>

  <table>
    <tr>
      <th style="text-align: center">6098_57837_000008_000000 (Ground-Truth) </th>
      <th style="text-align: center">6098_57837_000008_000000 (Synthesized) </th>
    </tr>
    <tr>
      <td style="text-align: center"><audio controls="controls">
          <source src="./demo/LibriTTS/6098_57837_000008_000000_ground-truth.wav" autoplay />
        </audio></td>
      <td style="text-align: center"><audio controls="controls">
          <source src="./demo/LibriTTS/6098_57837_000008_000000_synthesized.wav" autoplay />
        </audio></td>
    </tr>
  </table>
  <div><b>Speaker</b>: <em>6098</em></div>
  <div><b>Text</b>: <em>word was sent of their predicament to the nearest fort, and lieutenant pershing was sent with a
      small detachment to their rescue.</em></div>

  <hr>

  <table>
    <tr>
      <th style="text-align: center">6544_231862_000065_000001 (Ground-Truth) </th>
      <th style="text-align: center">6544_231862_000065_000001 (Synthesized) </th>
    </tr>
    <tr>
      <td style="text-align: center"><audio controls="controls">
          <source src="./demo/LibriTTS/6544_231862_000065_000001_ground-truth.wav" autoplay />
        </audio></td>
      <td style="text-align: center"><audio controls="controls">
          <source src="./demo/LibriTTS/6544_231862_000065_000001_synthesized.wav" autoplay />
        </audio></td>
    </tr>
  </table>
  <div><b>Speaker</b>: <em>6544</em></div>
  <div><b>Text</b>: <em>denzil did look, and uttered a second cry more startling than the first.</em></div>

  <hr>

  <table>
    <tr>
      <th style="text-align: center">968_122545_000053_000000 (Ground-Truth) </th>
      <th style="text-align: center">968_122545_000053_000000 (Synthesized) </th>
    </tr>
    <tr>
      <td style="text-align: center"><audio controls="controls">
          <source src="./demo/LibriTTS/968_122545_000053_000000_ground-truth.wav" autoplay />
        </audio></td>
      <td style="text-align: center"><audio controls="controls">
          <source src="./demo/LibriTTS/968_122545_000053_000000_synthesized.wav" autoplay />
        </audio></td>
    </tr>
  </table>
  <div><b>Speaker</b>: <em>968</em></div>
  <div><b>Text</b>: <em>the bee fought the window angrily, up and down, up and down, for several minutes; then found the
      open glass and whirled
      out into the sunshine, joyfully.</em></div>

  </p>

</body>

</html>