<?xml version="1.0" encoding="utf-8" standalone="yes"?><rss version="2.0" xmlns:atom="http://www.w3.org/2005/Atom"><channel><title>Selfhosting on ahenriksson.com</title><link>/tags/selfhosting/</link><description>Recent content in Selfhosting on ahenriksson.com</description><generator>Hugo</generator><language>en</language><lastBuildDate>Sun, 22 Feb 2026 22:10:00 +0100</lastBuildDate><atom:link href="/tags/selfhosting/index.xml" rel="self" type="application/rss+xml"/><item><title>ljudanteckning: I want to search my media library using text (open source, runs on multiple GPUs, using CUDA, written in modern Python)</title><link>/posts/make-your-media-library-searchable-by-text/</link><pubDate>Sun, 22 Feb 2026 22:10:00 +0100</pubDate><guid>/posts/make-your-media-library-searchable-by-text/</guid><description>Scan folders, split audio, transcribe in parallel across NVIDIA GPUs, then export subtitles + timestamped text next to each media file.</description></item></channel></rss>