<?xml version="1.0" encoding="UTF-8"?>
<rss version="2.0" xmlns:atom="http://www.w3.org/2005/Atom">
  <channel>
    <title>Audio &amp; Speech Models — FoundationIndex</title>
    <link>https://foundationindex.ustunozgur.com/s/audio-speech-models</link>
    <description>Foundation models for understanding and generating human speech, music, and sound—evolving from transcription (Whisper) and text-to-speech into real-time full-duplex voice conversations.</description>
    <language>en</language>
    <atom:link href="https://foundationindex.ustunozgur.com/s/audio-speech-models/feed.xml" rel="self" type="application/rss+xml" />
    <item>
      <title>Audio &amp; Speech Models · Paper · 2024-09-17</title>
      <link>https://foundationindex.ustunozgur.com/s/audio-speech-models#audio-speech-moshi-2024</link>
      <guid>https://foundationindex.ustunozgur.com/s/audio-speech-models#audio-speech-moshi-2024</guid>
      <pubDate>Tue, 17 Sep 2024 00:00:00 GMT</pubDate>
      <description>&lt;p&gt;Défossez et al. at Kyutai released Moshi, an open-weight full-duplex speech-to-speech foundation model using the Mimi neural audio codec to listen and speak simultaneously with ~200ms latency.&lt;/p&gt;&lt;p&gt;Sources: &lt;a href=&quot;https://arxiv.org/abs/2410.00037&quot;&gt;Moshi: a speech-text foundation model for real-time dialogue&lt;/a&gt;&lt;/p&gt;</description>
    </item>
    <item>
      <title>Audio &amp; Speech Models · Paper · 2023-01-05</title>
      <link>https://foundationindex.ustunozgur.com/s/audio-speech-models#audio-speech-valle-2023</link>
      <guid>https://foundationindex.ustunozgur.com/s/audio-speech-models#audio-speech-valle-2023</guid>
      <pubDate>Thu, 05 Jan 2023 00:00:00 GMT</pubDate>
      <description>&lt;p&gt;Wang et al. at Microsoft published VALL-E, framing text-to-speech as conditional neural audio codec language modeling—enabling zero-shot voice cloning from a 3-second acoustic prompt.&lt;/p&gt;&lt;p&gt;Sources: &lt;a href=&quot;https://arxiv.org/abs/2301.02111&quot;&gt;Neural Codec Language Models are Zero-Shot Text to Speech Synthesizers&lt;/a&gt;&lt;/p&gt;</description>
    </item>
    <item>
      <title>Audio &amp; Speech Models · Paper · 2022-12-06</title>
      <link>https://foundationindex.ustunozgur.com/s/audio-speech-models#audio-speech-whisper-2022</link>
      <guid>https://foundationindex.ustunozgur.com/s/audio-speech-models#audio-speech-whisper-2022</guid>
      <pubDate>Tue, 06 Dec 2022 00:00:00 GMT</pubDate>
      <description>&lt;p&gt;Radford et al. at OpenAI published Whisper, an open-weight encoder-decoder Transformer trained on 680,000 hours of weakly supervised multilingual audio that became the industry standard for speech recognition and transcription.&lt;/p&gt;&lt;p&gt;Sources: &lt;a href=&quot;https://arxiv.org/abs/2212.04356&quot;&gt;Robust Speech Recognition via Large-Scale Weak Supervision&lt;/a&gt;&lt;/p&gt;</description>
    </item>
  </channel>
</rss>
