<?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>Children&#039;s voices &#8211; Politepix</title>
		<atom:link href="/forums/topic/childrens-voices/feed/" rel="self" type="application/rss+xml" />
		<link>/forums/topic/childrens-voices/feed/</link>
		<description></description>
		<lastBuildDate>Tue, 23 Apr 2024 14:55:47 +0000</lastBuildDate>
		<generator>https://bbpress.org/?v=2.6.9</generator>
		<language>en-US</language>

		
														
					
				<item>
					<guid>/forums/topic/childrens-voices/#post-1017290</guid>
					<title><![CDATA[Children&#039;s voices]]></title>
					<link>/forums/topic/childrens-voices/#post-1017290</link>
					<pubDate>Thu, 23 May 2013 22:31:57 +0000</pubDate>
					<dc:creator>woodyard</dc:creator>

					<description>
						<![CDATA[
						<p>My app uses a dynamically created grammar of ten words.  Recognition for adult voices works well, but recognition of children&#8217;s voices is about 20% lower.</p>
<p>There is an old post regarding keyword spotting here that talks about the use of the kFILLPROB parameter:</p>
<p><a href="/forums/topic/add-key-word-spotting-to-openears-library/">/forums/topic/add-key-word-spotting-to-openears-library/</a></p>
<p>I&#8217;m wondering if adjusting this value might resolve the issue?</p>
						]]>
					</description>

					
					
				</item>

			
				<item>
					<guid>/forums/topic/childrens-voices/#post-1017293</guid>
					<title><![CDATA[Reply To: Children&#039;s voices]]></title>
					<link>/forums/topic/childrens-voices/#post-1017293</link>
					<pubDate>Fri, 24 May 2013 07:21:08 +0000</pubDate>
					<dc:creator>Halle Winkler</dc:creator>

					<description>
						<![CDATA[
						<p>Historically, people have had pretty good results by doing some adaptation using kid speech:</p>
<p><a href="/2012/11/19/openears-tip-and-tricks-3-acoustic-model-adaptation/">/2012/11/19/openears-tip-and-tricks-3-acoustic-model-adaptation/</a></p>
<p>I don&#8217;t know why raising or lowering the filler dictionary probability would affect recognition of speech for a different group of speakers but there&#8217;s no harm in trying it.</p>
						]]>
					</description>

					
					
				</item>

					
		
	</channel>
	</rss>

