<?xml version="1.0" encoding="utf-8" standalone="yes"?><rss version="2.0" xmlns:atom="http://www.w3.org/2005/Atom"><channel><title>OpenCouncil on Angelos Papamichail</title><link>https://angelospk.github.io/tags/opencouncil/</link><description>Recent content in OpenCouncil on Angelos Papamichail</description><generator>Hugo</generator><language>en</language><lastBuildDate>Tue, 21 Jul 2026 00:00:00 +0000</lastBuildDate><atom:link href="https://angelospk.github.io/tags/opencouncil/index.xml" rel="self" type="application/rss+xml"/><item><title>Teaching a machine to hear a Greek municipal council</title><link>https://angelospk.github.io/posts/greek-asr-finetuning/</link><pubDate>Tue, 21 Jul 2026 00:00:00 +0000</pubDate><guid>https://angelospk.github.io/posts/greek-asr-finetuning/</guid><description>&lt;p>&lt;em>This is an English adaptation of a post I originally wrote in Greek for the Schema Labs Substack. The original is here: &lt;a href="https://schemalabs.substack.com/p/asr-finetuning" class="external-link" target="_blank" rel="noopener">Μαθαίνοντας μια μηχανή να ακούει ένα ελληνικό δημοτικό συμβούλιο&lt;/a>.&lt;/em>&lt;/p>
&lt;hr>
&lt;p>OpenCouncil transcribes Greek municipal council meetings with ASR (Automatic Speech Recognition), and then someone fixes the AI&amp;rsquo;s mistakes by hand. The team spends several hours on each meeting to get those corrections right, so the final transcript matches the audio. That led us to a question: can we fine-tune an open ASR model on OpenCouncil&amp;rsquo;s own data and cut down the time spent on manual corrections? I&amp;rsquo;m Angelos, and this is the problem I&amp;rsquo;m working on for my Google Summer of Code project.&lt;/p></description></item></channel></rss>