<?xml version="1.0" encoding="utf-8" standalone="yes"?><rss version="2.0" xmlns:atom="http://www.w3.org/2005/Atom"><channel><title>Audio on Steve Murr</title><link>https://stevemurr.com/tags/audio/</link><description>Recent content in Audio on Steve Murr</description><generator>Hugo</generator><language>en-us</language><lastBuildDate>Wed, 08 Apr 2026 00:00:00 +0000</lastBuildDate><atom:link href="https://stevemurr.com/tags/audio/index.xml" rel="self" type="application/rss+xml"/><item><title>Training ML Models for Instrumental Music Generation</title><link>https://stevemurr.com/posts/music-generation/music-generation/</link><pubDate>Wed, 08 Apr 2026 00:00:00 +0000</pubDate><guid>https://stevemurr.com/posts/music-generation/music-generation/</guid><description>The most practical path to building an instrumental music generation system today is fine-tuning Meta&amp;rsquo;s MusicGen on 500–2,000 curated tracks — costing $100–300 in compute. This report covers architectures, datasets, training pipelines, compute costs, and actionable recommendations.</description></item></channel></rss>