<?xml version="1.0" encoding="utf-8"?>
<rss version="2.0"
    xmlns:dc="http://purl.org/dc/elements/1.1/"
     xmlns:admin="http://webns.net/mvcb/"
     xmlns:content="http://purl.org/rss/1.0/modules/content/"
     xmlns:rdf="http://www.w3.org/1999/02/22-rdf-syntax-ns#">
	<channel>
	<title>MetaFilter posts tagged with speech and technology</title>
	<link>http://www.metafilter.com/tags/speech+technology</link>
	<description>Posts tagged with 'speech' and 'technology' at MetaFilter.</description>
	<pubDate>Tue, 02 Mar 2010 16:32:43 -0800</pubDate> <lastBuildDate>Tue, 02 Mar 2010 16:32:43 -0800</lastBuildDate>

	<language>en-us</language>
	<docs>http://blogs.law.harvard.edu/tech/rss</docs>
	<ttl>60</ttl>
	<item>
		<title>It takes two to speak truth: One to speak and another to hear</title>
		<link>http://www.metafilter.com/89705/It%2Dtakes%2Dtwo%2Dto%2Dspeak%2Dtruth%2DOne%2Dto%2Dspeak%2Dand%2Danother%2Dto%2Dhear</link>
		<description> &lt;a href=&quot;http://videogum.com/148961/roger-ebert-the-best-finally-gets-his-voice-back/yay/&quot;&gt;Roger Ebert gets his voice back&lt;/a&gt; The movie critic whose ongoing fight with cancer and loss of speech has been &lt;a href=&quot;http://www.metafilter.com/89255/When-I-am-writing-my-problems-become-invisible-and-I-am-the-same-person-I-always-was-All-is-well-I-am-as-I-should-be&quot;&gt;profiled&lt;/a&gt; has now gained his voice back through the efforts of Scotland-based &lt;a href=&quot;http://www.cereproc.com/about&quot;&gt;CereProc&lt;/a&gt;. 

In an &lt;a href=&quot;http://www.npr.org/templates/story/story.php?storyId=124087291&quot;&gt;NPR interview&lt;/a&gt;, CereProc&apos;s Dr. Matthew Aylett explains how Ebert&apos;s voice was reconstructed through phoneme mining, sifting through commentaries and other audio recordings from his body of work. </description>
		<guid isPermaLink="false">tag:metafilter.com,2010:site.89705</guid>
		<pubDate>Tue, 02 Mar 2010 16:32:43 -0800</pubDate>
		<category>cereproc</category>
		<category>datamining</category>
		<category>ebert</category>
		<category>mining</category>
		<category>npr</category>
		<category>phoneme</category>
		<category>rogerebert</category>
		<category>scotland</category>
		<category>speech</category>
		<category>technology</category>
		<dc:creator>Blazecock Pileon</dc:creator>
	</item>
      <item>
		<title>Hitler Speaks</title>
		<link>http://www.metafilter.com/70124/Hitler%2DSpeaks</link>
		<description> &lt;a href=&quot;http://video.google.com/videoplay?docid=2763127556620650689&amp;q=hitler+speaks+duration%3Along&amp;total=36&amp;start=0&amp;num=10&amp;so=0&amp;type=search&amp;plindex=0&quot;&gt;Hitler Speaks&lt;/a&gt;
&lt;br&gt;
&lt;br&gt;
Using advanced speech recognition technology, researchers and voice-over actors have been able to put a soundtrack to long-silent video relics of Adolf Hitler: &lt;a href=&quot;http://www.thirdreichruins.com/eva_movies.htm&quot;&gt;Eva Braun&apos;s infamous home movies&lt;/a&gt; filmed at the &lt;a href=&quot;http://en.wikipedia.org/wiki/Berghof_%28Hitler%29&quot;&gt;Berghof&lt;/a&gt;, private filmed meetings between Hitler and various Reich cronies, as well as the last known footage of him taped before an awkward bunch of Hitler Youth at the Reichstag in the final days of the war made famous in &lt;a href=&quot;http://imdb.com/title/tt0363163/&quot;&gt;Downfall&lt;/a&gt;.  Chilling stuff.
&lt;br&gt;
&lt;br&gt;
&lt;a href=&quot;http://digg.com/educational/Hitler_Speaks_Documentary_about_Hitler_s_Private_Videos&quot;&gt;Via&lt;/a&gt;.  </description>
		<guid isPermaLink="false">tag:metafilter.com,2008:site.70124</guid>
		<pubDate>Sat, 22 Mar 2008 14:39:37 -0800</pubDate>
		<category>history</category>
		<category>hitler</category>
		<category>speech</category>
		<category>technology</category>
		<category>video</category>
		<dc:creator>auralcoral</dc:creator>
	</item>
      <item>
		<title>Search YouTube</title>
		<link>http://www.metafilter.com/57483/Search%2DYouTube</link>
		<description>&lt;a href="http://podzinger.com/"&gt;Podzinger now lets you search through spoken words on YouTube.&lt;/a&gt; Podzinger has long done speech recognition-based searches of podcasts, including neat features like excerpting relevant bits of the podcast, but the YouTube search is new, and still in its infancy. Podzinger comes from BBN, one of the creators of the &lt;a href=&quot;http://www.livinginternet.com/i/ii_arpanet.htm&quot;&gt;internet &lt;/a&gt;and &lt;a href=&quot;http://openmap.bbn.com/~tomlinso/ray/home.html&quot;&gt;email&lt;/a&gt;,  and which was the setting for one of the more &lt;a href=&quot;http://www.stanford.edu/group/SHR/4-2/text/dialogues.html&quot;&gt;humorous incidents in AI history&lt;/a&gt; &lt;small&gt;[scroll down to &quot;an accidental conversation&quot;]&lt;/small&gt;.  </description>
		<guid isPermaLink="false">tag:metafilter.com,2007:site.57483</guid>
		<pubDate>Thu, 04 Jan 2007 07:54:04 -0800</pubDate>
		<category>podzinger</category>
		<category>speech</category>
		<category>speechrecognition</category>
		<category>technology</category>
		<dc:creator>blahblahblah</dc:creator>
	</item>
      <item>
		<title>Evacuation Plans</title>
		<link>http://www.metafilter.com/27803/Evacuation%2DPlans</link>
		<description>&lt;a href="http://www.wired.com/news/conflict/0,2100,60144,00.html"&gt;Writing a story is a felony violation in Oklahoma.&lt;/a&gt; High school student who wrote a story about armed invasion of his school faces jail time. Prosectors admit the law is Orwellian, but need a test-case. The story itself is derived from &lt;a href=&quot;http://www2.una.edu/jburney/JI%20AD%20COPY.TXT&quot;&gt;sample text&lt;/a&gt; included with an Adobe Pagemaker tutorial, describing hurrican evacuation instructions.  </description>
		<guid isPermaLink="false">tag:metafilter.com,2003:site.27803</guid>
		<pubDate>Fri, 22 Aug 2003 07:58:02 -0800</pubDate>
		<category>aclu</category>
		<category>highschool</category>
		<category>oklahoma</category>
		<category>speech</category>
		<category>technology</category>
		<dc:creator>swift</dc:creator>
	</item>
      
	</channel>
</rss>


