<?xml version="1.0" encoding="UTF-8"?><rss version="2.0"
	xmlns:content="http://purl.org/rss/1.0/modules/content/"
	xmlns:dc="http://purl.org/dc/elements/1.1/"
	xmlns:atom="http://www.w3.org/2005/Atom"
	>
<channel>
	<title>Comments on: Speaking Freely</title>
	<atom:link href="http://www.loudmouthman.com/2008/05/06/speaking-freely/feed/" rel="self" type="application/rss+xml" />
	<link>http://www.loudmouthman.com/2008/05/06/speaking-freely/</link>
	<description>Social Networker and Opinoneer</description>
	<pubDate>Fri, 25 Jul 2008 17:52:38 +0000</pubDate>
	<generator>http://wordpress.org/?v=2.5</generator>
		<item>
		<title>By: telecom lately2</title>
		<link>http://www.loudmouthman.com/2008/05/06/speaking-freely/#comment-21615</link>
		<dc:creator>telecom lately2</dc:creator>
		<pubDate>Tue, 06 May 2008 15:25:59 +0000</pubDate>
		<guid isPermaLink="false">http://www.loudmouthman.com/?p=197#comment-21615</guid>
		<description>I agree, and your idea's good.  

But what do you mean by "accuracy"?  If you read Spinvox material, an unidentified percentage of the messages spoken into their system are manually transcribed.  There's no data about how much their recognition system "recognized", and how much some typists typed.  Its a fun idea, but I can't see it as being technologically practical.</description>
		<content:encoded><![CDATA[<p>I agree, and your idea&#8217;s good.  </p>
<p>But what do you mean by &#8220;accuracy&#8221;?  If you read Spinvox material, an unidentified percentage of the messages spoken into their system are manually transcribed.  There&#8217;s no data about how much their recognition system &#8220;recognized&#8221;, and how much some typists typed.  Its a fun idea, but I can&#8217;t see it as being technologically practical.</p>
]]></content:encoded>
	</item>
</channel>
</rss>
