feat: audio native

elevenlabs · Dec 22, 2024 · 88a6b38 · 88a6b38
1 parent 814ee69
commit 88a6b38
Show file tree

Hide file tree

Showing 3 changed files with 78 additions and 9 deletions.
diff --git a/fern/assets/styles/globals.css b/fern/assets/styles/globals.css
@@ -7,3 +7,13 @@
 .fern-sidebar-icon {
   display: none !important;
 }
+
+/* Hide first HR in changelog */
+.fern-changelog > main > section + hr {
+  display: none;
+}
+
+/* Reduce padding on first changelog entry */
+.fern-changelog > main > section.fern-changelog-entry {
+  padding-bottom: 0px;
+}
diff --git a/fern/docs.yml b/fern/docs.yml
@@ -141,6 +141,8 @@ js:
     strategy: lazyOnload
   - path: assets/scripts/audio-player.js
     strategy: lazyOnload
+  - url: https://elevenlabs.io/player/audioNativeHelper.js
+    strategy: lazyOnload
 
 analytics:
   posthog:

diff --git a/fern/docs/pages/guides/text-to-speech.mdx b/fern/docs/pages/guides/text-to-speech.mdx
@@ -5,13 +5,33 @@ subtitle: 'Learn how to turn text into human-like spoken audio with ElevenLabs.'
 
 ## Overview
 
-ElevenLabs Text to Speech (TTS) API turns text into natural-sounding audio with human-like intonation, pacing, and contextual understanding. Our system supports many languages, multiple voice styles, and real-time streaming. Whether you’re creating audiobooks, localizing video content, or building an interactive voice application, ElevenLabs provides the tools and speed you need.
+<div
+  id="elevenlabs-audionative-widget"
+  data-height="90"
+  data-width="100%"
+  data-frameborder="no"
+  data-scrolling="no"
+  data-publicuserid="b8723bd5176782a2716d531124554c58dace5eb6b7d831c2ceb7ce48a945c64f"
+  data-playerurl="https://elevenlabs.io/player/index.html"
+>
+  Loading the{' '}
+  <a href="https://elevenlabs.io/text-to-speech" target="_blank" rel="noopener">
+    Elevenlabs Text to Speech
+  </a>{' '}
+  AudioNative Player...
+</div>
+
+<script src="https://elevenlabs.io/player/audioNativeHelper.js" type="text/javascript"></script>
+
+ElevenLabs Text to Speech (TTS) API transforms text into lifelike audio with nuanced intonation, pacing and emotional awareness. [Our models](/docs/models) smoothly adapt to textual cues across 32 languages and multiple voice styles. Ideal for everything from global brand campaigns to immersive entertainment.
 
 <elevenlabs-audio-player
     audio-title="George"
     audio-src="https://storage.googleapis.com/eleven-public-cdn/audio/marketing/george.mp3"
 />
 
+Explore our [Voice Library](https://elevenlabs.io/community) to find the perfect voice for your project.
+
 ## Quickstart
 
 The main TTS endpoint requires:
@@ -77,7 +97,7 @@ Below is an example showing how to make a basic request to create an MP3 file.
   </Tab>
 </Tabs>
 
-Run your function, and you’ll have a brand-new MP3 file containing your AI voiceover.
+Run your function, and you'll have a brand-new MP3 file containing your AI voiceover.
 
 <CardGroup cols={2}>
   <Card title="Audio Quality" icon="music">
@@ -92,7 +112,7 @@ Run your function, and you’ll have a brand-new MP3 file containing your AI voi
     ElevenLabs supports more than 3,000 voices across 32 languages, including:
     <ul>
       <li>Default curated voices</li>
-      <li>Voice Library of shared Clones</li>
+      <li><a href="https://elevenlabs.io/community">Voice Library</a> of shared Clones</li>
       <li>Instant Voice Cloning</li>
       <li>Voice Design for new voices</li>
     </ul>
@@ -101,7 +121,7 @@ Run your function, and you’ll have a brand-new MP3 file containing your AI voi
 
 <CardGroup cols={2}>
   <Card title="Streaming Real-Time Audio" icon="circle-play">
-    For interactive or time-sensitive applications, stream audio as it’s being generated. This reduces wait times, so users can hear the speech almost instantly.
+    For interactive or time-sensitive applications, stream audio as it's being generated. This reduces wait times, so users can hear the speech almost instantly.
   </Card>
 
   <Card title="Output Formats" icon="file-audio">
@@ -165,9 +185,46 @@ Run your function, and you’ll have a brand-new MP3 file containing your AI voi
 
 ## Supported languages
 
-ElevenLabs TTS offers full or partial support for 32 languages:
-• English • Spanish • French • Italian • Polish • Portuguese • Russian • German • Japanese • Korean • Chinese (simplified) • Hindi • Arabic, and more!  
-Choose a “multilingual” model family for the broadest coverage.
+ElevenLabs TTS supports 32 languages with our multilingual models, enabling global reach and natural-sounding localization:
+
+<CardGroup cols={3}>
+  <Card title="European">
+    • English
+    • French
+    • German
+    • Italian
+    • Spanish
+    • Portuguese
+    • Polish
+    • Dutch
+    • Swedish
+    • Norwegian
+    • Greek
+    • Romanian
+    • Hungarian
+    • Bulgarian
+    • Croatian
+    • Czech
+    • Danish
+    • Finnish
+    • Slovak
+    • Ukrainian
+  </Card>
+
+{' '}
+
+<Card title="Asian">
+  • Chinese • Japanese • Korean • Vietnamese • Tamil • Hindi • Filipino • Malay • Indonesian
+</Card>
+
+  <Card title="Other">
+    • Arabic
+    • Russian
+    • Turkish
+  </Card>
+</CardGroup>
+
+Choose a "multilingual" model family (like `eleven_multilingual_v2`) for the broadest language coverage. All our AI voices can speak any of these languages, making it easy to create consistent brand voices across multiple territories.
 
 ## FAQ
 
@@ -176,7 +233,7 @@ Choose a “multilingual” model family for the broadest coverage.
     Absolutely. Our voice settings include parameters like Stability, Similarity, and Style
     Exaggeration. Keep stability high for steadier speech, lower it for more emotional variance.
   </Accordion>
-  <Accordion title="Can I clone my own voice or a specific speaker’s voice?">
+  <Accordion title="Can I clone my own voice or a specific speaker's voice?">
     Yes. Instant Voice Cloning quickly mimics another speaker from short clips. For high-fidelity
     clones, check out our Professional Voice Clone with advanced training on larger datasets.
   </Accordion>
@@ -192,7 +249,7 @@ Choose a “multilingual” model family for the broadest coverage.
     The TTS engine can be nondeterministic. For consistency, use the optional seed parameter, though
     subtle differences may still occur.
   </Accordion>
-  <Accordion title="What’s the best practice for large text conversions?">
+  <Accordion title="What's the best practice for large text conversions?">
     Split long text into segments and use streaming for real-time playback or partial saving to
     disk. For smoother transitions, leverage "previous_text" or "previous_request_ids".
   </Accordion>