<?xml version="1.0" encoding="utf-8" standalone="yes"?>
<rss version="2.0" xmlns:atom="http://www.w3.org/2005/Atom" xmlns:content="http://purl.org/rss/1.0/modules/content/">
  <channel>
    <title>语音对话系统， on 语音/音频论文速递</title>
    <link>https://nanless.github.io/audio-paper-digest-blog/tags/%E8%AF%AD%E9%9F%B3%E5%AF%B9%E8%AF%9D%E7%B3%BB%E7%BB%9F/</link>
    <description>Recent content in 语音对话系统， on 语音/音频论文速递</description>
    <generator>Hugo</generator>
    <language>zh-cn</language>
    <lastBuildDate>Mon, 20 Apr 2026 00:00:00 +0000</lastBuildDate>
    <atom:link href="https://nanless.github.io/audio-paper-digest-blog/tags/%E8%AF%AD%E9%9F%B3%E5%AF%B9%E8%AF%9D%E7%B3%BB%E7%BB%9F/index.xml" rel="self" type="application/rss+xml" />
    <item>
      <title>Qwen3.5-Omni Technical Report</title>
      <link>https://nanless.github.io/audio-paper-digest-blog/posts/2026-04-20-qwen35-omni-technical-report/</link>
      <pubDate>Mon, 20 Apr 2026 00:00:00 +0000</pubDate>
      <guid>https://nanless.github.io/audio-paper-digest-blog/posts/2026-04-20-qwen35-omni-technical-report/</guid>
      <description>Qwen3.5-Omni 是一个旨在统一理解、推理、生成与行动的全模态大语言模型。它**解决**了现有模型在实时交互、长上下文音视频处理、流式语音生成稳定性以及多语言支持等方面的局限性。**方法上**，它基于Thinker-Talker架构，引入了Hybrid MoE以提升效率，采用显式时间戳替代稀</description>
    </item>
  </channel>
</rss>
