<?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>Speech-Enhancement on Cognaptus</title>
    <link>https://cognaptus.com/tags/speech-enhancement/</link>
    <description>Recent content in Speech-Enhancement on Cognaptus</description>
    <generator>Hugo -- 0.145.0</generator>
    <language>en-us</language>
    <lastBuildDate>Fri, 09 Oct 2026 00:00:00 +0000</lastBuildDate>
    <atom:link href="https://cognaptus.com/tags/speech-enhancement/index.xml" rel="self" type="application/rss+xml" />
    <item>
      <title>Clean Speech Is a Prior, Not a Finish Line</title>
      <link>https://cognaptus.com/blog/2026-10-09-clean-speech-is-a-prior-not-a-finish-line/</link>
      <pubDate>Fri, 09 Oct 2026 00:00:00 +0000</pubDate>
      <guid>https://cognaptus.com/blog/2026-10-09-clean-speech-is-a-prior-not-a-finish-line/</guid>
      <description>A test-time adaptation method for speech enhancement shows that deployment-time correction works best when it is selective, input-dependent, and stopped before the clean-speech prior overwhelms the observed audio.</description>
    </item>
  </channel>
</rss>
