<?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>音声認識 on Studio Notes</title>
    <link>https://kingsworksub-jpg.github.io/tags/%E9%9F%B3%E5%A3%B0%E8%AA%8D%E8%AD%98/</link>
    <description>Recent content in 音声認識 on Studio Notes</description>
    <image>
      <title>Studio Notes</title>
      <url>https://kingsworksub-jpg.github.io/images/og/site-default.jpg</url>
      <link>https://kingsworksub-jpg.github.io/images/og/site-default.jpg</link>
    </image>
    <generator>Hugo</generator>
    <language>ja</language>
    <lastBuildDate>Tue, 13 Oct 2026 07:30:00 +0900</lastBuildDate>
    <atom:link href="https://kingsworksub-jpg.github.io/tags/%E9%9F%B3%E5%A3%B0%E8%AA%8D%E8%AD%98/index.xml" rel="self" type="application/rss+xml" />
    <item>
      <title>Whisperの日本語文字起こしはどのサイズから実用になるか、tinyからlarge-v3-turboまで</title>
      <link>https://kingsworksub-jpg.github.io/posts/whisper-model-size-japanese-benchmark-2026/</link>
      <pubDate>Tue, 13 Oct 2026 07:30:00 +0900</pubDate>
      <guid>https://kingsworksub-jpg.github.io/posts/whisper-model-size-japanese-benchmark-2026/</guid>
      <description>Whisperのtinyからlarge-v3、large-v3-turboまでを、日本語の文字誤り率と必要なVRAM、速度で並べた検証メモ。faster-whisperとwhisper.cppの差、日本語で起きやすい誤認識の対策もまとめた。</description>
    </item>
  </channel>
</rss>
