<?xml version="1.0" encoding="UTF-8"?>
<rss version="2.0" xmlns:atom="http://www.w3.org/2005/Atom">
<channel>
  <title>MATRIX · DiarizationLM</title>
  <link>https://derekwang2002.github.io/ai-intelligence/en/projects/diarizationlm/</link>
  <description>Material changes in AI models, agents, engineering and infrastructure, with primary sources and recommendations.</description>
  <language>en</language>
  <lastBuildDate>Mon, 02 Nov 2026 16:56:00 GMT</lastBuildDate>
  <atom:link href="https://derekwang2002.github.io/ai-intelligence/en/feeds/projects/diarizationlm.xml" rel="self" type="application/rss+xml"/>
  <item>
    <title>Google opens a 4B model that repairs speaker labels in transcripts</title>
    <link>https://derekwang2002.github.io/ai-intelligence/en/events/ev-20261004-01/</link>
    <guid isPermaLink="false">ev-20261004-01-published</guid>
    <pubDate>Sun, 04 Oct 2026 16:02:19 GMT</pubDate>
    <category>New</category>
    <category>TRIAL</category>
    <description>&lt;p&gt;Google released DiarizationLM-Gemma-4-E4B-v1 under Apache-2.0. The 4B model post-processes ASR and diarization output, with BF16 and two 4-bit GGUF variants for local use. Author tests report lower word diarization error and cpWER on Fisher, Callhome, ICSI and AMI, although the meeting test sets are small.&lt;/p&gt;&lt;p&gt;&lt;strong&gt;Why read: &lt;/strong&gt;DiarizationLM local GGUF files make speaker correction testable in one day.&lt;/p&gt;&lt;p&gt;Recommendation: TRIAL&lt;/p&gt;&lt;p&gt;Sources: &lt;a href=&quot;https://huggingface.co/google/DiarizationLM-Gemma-4-E4B-v1&quot;&gt;DiarizationLM-Gemma-4-E4B-v1&lt;/a&gt; · &lt;a href=&quot;https://github.com/google/speaker-id/tree/master/DiarizationLM&quot;&gt;Google DiarizationLM implementation&lt;/a&gt;&lt;/p&gt;</description>
  </item>
</channel>
</rss>
