<?xml version="1.0" encoding="UTF-8"?>
	<rss version="2.0"
		xmlns:content="http://purl.org/rss/1.0/modules/content/"
		xmlns:wfw="http://wellformedweb.org/CommentAPI/"
		xmlns:dc="http://purl.org/dc/elements/1.1/"
		xmlns:atom="http://www.w3.org/2005/Atom"

			>

	<channel>

		<title>How does recognition score works? &#8211; Politepix</title>
		<atom:link href="/forums/topic/how-does-recognition-score-works/feed/" rel="self" type="application/rss+xml" />
		<link>/forums/topic/how-does-recognition-score-works/feed/</link>
		<description></description>
		<lastBuildDate>Tue, 23 Apr 2024 14:39:51 +0000</lastBuildDate>
		<generator>https://bbpress.org/?v=2.6.9</generator>
		<language>en-US</language>

		
														
					
				<item>
					<guid>/forums/topic/how-does-recognition-score-works/#post-10201</guid>
					<title><![CDATA[How does recognition score works?]]></title>
					<link>/forums/topic/how-does-recognition-score-works/#post-10201</link>
					<pubDate>Fri, 06 Jul 2012 13:26:55 +0000</pubDate>
					<dc:creator>luiso1979</dc:creator>

					<description>
						<![CDATA[
						<p>Hello,</p>
<p>I&#8217;m not an expert in Speech Recognition. Could someone please explain to me how the recognition score works. What is the correspondence between the score and the reliability of the recognition? Which  is the minimum score that I should consider to ignore the hypothesis?</p>
<p>Thanks in advance</p>
<p>Luis</p>
						]]>
					</description>

					
					
				</item>

			
				<item>
					<guid>/forums/topic/how-does-recognition-score-works/#post-10204</guid>
					<title><![CDATA[Reply To: How does recognition score works?]]></title>
					<link>/forums/topic/how-does-recognition-score-works/#post-10204</link>
					<pubDate>Fri, 06 Jul 2012 14:47:29 +0000</pubDate>
					<dc:creator>Halle Winkler</dc:creator>

					<description>
						<![CDATA[
						<p>Hi Luis,</p>
<p>It&#8217;s very much relative to that particular speaker and session and the size of the language model. It can&#8217;t be reliably used to create arbitrary cutoffs in my experience except perhaps with extremely low values (like -500000 or lower). I wish it were more useful for the task of evaluating accuracy but I haven&#8217;t encountered a case yet where it was possible to rely on the score so I&#8217;ve reversed my old advice that it could be used in this way.</p>
						]]>
					</description>

					
					
				</item>

					
		
	</channel>
	</rss>

