<?xml version="1.0" encoding="utf-8" standalone="yes"?><rss version="2.0" xmlns:atom="http://www.w3.org/2005/Atom"><channel><title>WebSocket · Afa'</title><link>https://afa.demanou.com/tags/websocket/</link><description>WebSocket · A personal engineering notebook about software, systems, infrastructure, AI and things worth understanding.</description><language>en</language><copyright>© 2026 Afa'</copyright><managingEditor>Sébastien Demanou</managingEditor><generator>Hugo</generator><lastBuildDate>Sun, 04 Oct 2026 01:12:41 +0000</lastBuildDate><atom:link href="https://afa.demanou.com/tags/websocket/index.xml" rel="self" type="application/rss+xml"/><item><title>Oremi Ohunerin 4.0.0.b11: Turning Measurements into a Faster and More Reliable Audio Server</title><link>https://afa.demanou.com/posts/2026/oremi-ohunerin-4-0-0-b11-turning-measurements-into-a-faster-and-more-reliable-audio-server/</link><guid isPermaLink="true">https://afa.demanou.com/posts/2026/oremi-ohunerin-4-0-0-b11-turning-measurements-into-a-faster-and-more-reliable-audio-server/</guid><pubDate>Sat, 03 Oct 2026 00:00:00 +0000</pubDate><category>Engineering, Audio Processing, Wake Word, PocketSphinx, YAMNet, TensorFlow Lite, WebSocket, Performance, Benchmarking, Python, Oremi</category><description>A benchmark-driven look at Oremi Ohunerin 4.0.0.b11, covering audio inference performance, WebSocket concurrency, wake-word reliability, decoder isolation, and …</description></item></channel></rss>