<?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 AI on AI Tech Blog</title>
    <link>https://jesamkim.github.io/ai-tech-blog/categories/speech-ai/</link>
    <description>Recent content in Speech AI on AI Tech Blog</description>
    <generator>Hugo -- 0.147.6</generator>
    <language>ko</language>
    <lastBuildDate>Tue, 09 Jun 2026 20:00:00 +0900</lastBuildDate>
    <atom:link href="https://jesamkim.github.io/ai-tech-blog/categories/speech-ai/index.xml" rel="self" type="application/rss+xml" />
    <item>
      <title>음성을 이해하고 만들기까지: SSL부터 Zero-Shot Voice Cloning으로 가는 길</title>
      <link>https://jesamkim.github.io/ai-tech-blog/posts/2026-06-09-modern-speech-ai-from-ssl-to-zero-shot-voice-cloning/</link>
      <pubDate>Tue, 09 Jun 2026 20:00:00 +0900</pubDate>
      <guid>https://jesamkim.github.io/ai-tech-blog/posts/2026-06-09-modern-speech-ai-from-ssl-to-zero-shot-voice-cloning/</guid>
      <description>라벨 없이 음성을 학습하는 표현 학습(SSL)과 텍스트-음성 합성(TTS)이 어떻게 발전했고, 두 흐름이 zero-shot voice cloning에서 어떻게 만나는지를 정리했습니다. CPC와 wav2vec 2.0, HuBERT의 설계 결정부터 WaveNet 계열 보코더, Voicebox의 flow matching, CosyVoice 2의 LLM 기반 합성까지 직관 위주로 따라가 봅니다.</description>
    </item>
  </channel>
</rss>
