File size: 3,241 Bytes
cfaa94b
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
c45e1b5
cfaa94b
 
 
 
 
 
 
 
 
 
 
 
 
 
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
<!DOCTYPE html>
<html lang="en">
<head>
  <meta charset="utf-8">
  <meta name="description" content="Compare speech generated by PromptTTSD and other models across style control and dialogue generation tasks.">
  <meta name="keywords" content="PromptDialog, PromptTTSD, text-to-speech, dialogue, audio demo">
  <meta name="viewport" content="width=device-width, initial-scale=1">
  <title>PromptTTSD · Model demos</title>

  <link rel="stylesheet" href="./static/css/bulma.min.css">
  <link rel="stylesheet" href="./static/css/index.css?v=e05023061b9b">
</head>
<body>
  <a class="skip-link" href="#audio-demo">Skip to audio demos</a>
  <header class="site-header">
    <div class="container wide-container header-inner">
      <a class="wordmark" href="./model.html"><span class="brand-symbol" aria-hidden="true">p<span>⌁</span></span>PromptTTSD<span class="brand-dot">.</span></a>
      <nav aria-label="Main navigation">
        <a href="./">Dataset</a>
        <a href="./model.html" aria-current="page">Model demos</a>
      </nav>
    </div>
  </header>
  <main>
    <section class="page-intro demo-section" id="audio-demo">
      <div class="container wide-container">
        <div class="model-intro">
          <div class="hero-topline"><span class="eyebrow"><span class="status-dot"></span> MODEL DEMOS</span></div>
          <h1 class="publication-title">PromptTTSD</h1>
          <h2 class="hero-subtitle">Controllable Text-to-Spoken Dialogue Synthesis<br>with <em>Open-Ended Text Descriptions</em></h2>
          <p class="hero-description">We trained PromptTTSD by fine-tuning SoulX-Podcast on the PromptDialog dataset. To the best of our knowledge, PromptTTSD is the first open-source spoken dialogue synthesis model to support both zero-shot voice cloning and natural-language style control.</p>
          <div class="hero-actions">
            <a class="button-secondary brand-link" href="https://github.com/bigai-ai/PromptTTSD" target="_blank" rel="noopener"><img src="./static/icons/github.svg" width="19" height="19" alt=""> Code on GitHub <span aria-hidden="true">↗</span></a>
            <a class="button-secondary brand-link" href="https://huggingface.co/alphappp/PromptTTSD/tree/main" target="_blank" rel="noopener"><img src="./static/icons/huggingface.svg" width="21" height="20" alt=""> Model weights <span aria-hidden="true">↗</span></a>
            <a class="button-secondary brand-link" role="link" aria-disabled="true"><img src="./static/icons/arxiv.svg" width="19" height="19" alt=""> Read the paper <span aria-hidden="true">↗</span></a>
          </div>
        </div>
        <nav class="task-directory" id="task-directory" aria-label="Audio demo tasks"><p class="loading">Loading tasks…</p></nav>
        <div class="task-list" id="task-list" aria-live="polite"><div class="task-card"><p class="loading">Loading audio demos…</p></div></div>
      </div>
    </section>
  </main>
  <footer class="site-footer"><div class="container wide-container"><a class="wordmark" href="#">PromptTTSD<span class="brand-dot">.</span></a><a href="./">PromptDialog dataset</a><a href="#audio-demo">Back to top ↑</a></div></footer>

  <script type="module" src="./static/js/model.js"></script>
</body>
</html>