<?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>Hypothesis recognition scores &#8211; Politepix</title>
		<atom:link href="/forums/topic/hypothesis-recognition-scores/feed/" rel="self" type="application/rss+xml" />
		<link>/forums/topic/hypothesis-recognition-scores/feed/</link>
		<description></description>
		<lastBuildDate>Tue, 23 Apr 2024 14:55:03 +0000</lastBuildDate>
		<generator>https://bbpress.org/?v=2.6.9</generator>
		<language>en-US</language>

		
														
					
				<item>
					<guid>/forums/topic/hypothesis-recognition-scores/#post-1016921</guid>
					<title><![CDATA[Hypothesis recognition scores]]></title>
					<link>/forums/topic/hypothesis-recognition-scores/#post-1016921</link>
					<pubDate>Thu, 11 Apr 2013 02:36:20 +0000</pubDate>
					<dc:creator>uzziel</dc:creator>

					<description>
						<![CDATA[
						<p>Halle,</p>
<p>One more question.  When pocketsphinxDidReceiveHypothesis and pocketsphinxDidReceiveNBestHypothesisArray return, they pass along a hypothesis or hypothoses along with a recognition score.</p>
<p>What exactly does the recognition score mean?  I&#8217;ve seen values from 0 to -8000; I am assuming that a cardinally larger score is a worse match?  e.g., -10 is a more certain match than -100 &#8211; is that correct?</p>
						]]>
					</description>

					
					
				</item>

			
				<item>
					<guid>/forums/topic/hypothesis-recognition-scores/#post-1016924</guid>
					<title><![CDATA[Reply To: Hypothesis recognition scores]]></title>
					<link>/forums/topic/hypothesis-recognition-scores/#post-1016924</link>
					<pubDate>Thu, 11 Apr 2013 06:48:12 +0000</pubDate>
					<dc:creator>Halle Winkler</dc:creator>

					<description>
						<![CDATA[
						<p>Correct, they are probabilities with zero being certainty. They don&#8217;t have an enormous amount of utility (they are very subject to environmental conditions so you can&#8217;t use any kind of objective cutoff from session to session in order to say that a probability is too low) but you can compare them to each other within a single session, for instance.</p>
						]]>
					</description>

					
					
				</item>

					
		
	</channel>
	</rss>

