<?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>Comments on: Neural networks learning how to talk</title>
	<link>http://ask.metafilter.com/56945/Neural-networks-learning-how-to-talk/</link>
	<description>Comments on Ask MetaFilter post Neural networks learning how to talk</description>
	<pubDate>Tue, 13 Feb 2007 16:17:24 -0800</pubDate>
	<lastBuildDate>Tue, 13 Feb 2007 16:17:24 -0800</lastBuildDate>
	<language>en-us</language>
	<docs>http://blogs.law.harvard.edu/tech/rss</docs>
	<ttl>60</ttl>

	<item>
		<title>Question: Neural networks learning how to talk</title>
		<link>http://ask.metafilter.com/56945/Neural-networks-learning-how-to-talk</link>	
		<description>I&apos;m looking for a mp3 of a computer learning how to talk. &lt;br /&gt;&lt;br /&gt; My CS professor played us a clip of a neural network learning how to turn text into speech.  It starts out just babbling and eventually figures out stuff like spaces between words and so on. I think it&apos;s being trained on another synthesizer, and was working at the phonetic level (rather than outputting PCM data).&lt;br&gt;
&lt;br&gt;
I sort of think it might have been a Bell Labs project?</description>
		<guid isPermaLink="false">post:ask.metafilter.com,2007:site.56945</guid>
		<pubDate>Tue, 13 Feb 2007 16:07:22 -0800</pubDate>
		<dc:creator>aubilenon</dc:creator>
		
			<category>speech</category>
		
			<category>ai</category>
		
			<category>resolved</category>
		
	</item> <item>
		<title>By: logicpunk</title>
		<link>http://ask.metafilter.com/56945/Neural-networks-learning-how-to-talk#856114</link>	
		<description>It was probably Sejnowski&apos;s &lt;a href=&quot;http://www.cnl.salk.edu/ParallelNetsPronounce/nettalk.mp3&quot;&gt;NETtalk.&lt;/a&gt;</description>
		<guid isPermaLink="false">comment:ask.metafilter.com,2007:site.56945-856114</guid>
		<pubDate>Tue, 13 Feb 2007 16:17:24 -0800</pubDate>
		<dc:creator>logicpunk</dc:creator>
	</item><item>
		<title>By: aubilenon</title>
		<link>http://ask.metafilter.com/56945/Neural-networks-learning-how-to-talk#856244</link>	
		<description>Yes!  That is it!  Hooray for The Internet!  Hooray for logicpunk!</description>
		<guid isPermaLink="false">comment:ask.metafilter.com,2007:site.56945-856244</guid>
		<pubDate>Tue, 13 Feb 2007 18:11:49 -0800</pubDate>
		<dc:creator>aubilenon</dc:creator>
	</item><item>
		<title>By: blag</title>
		<link>http://ask.metafilter.com/56945/Neural-networks-learning-how-to-talk#856752</link>	
		<description>That&apos;s awesome. It sounds like Aphex Twin.</description>
		<guid isPermaLink="false">comment:ask.metafilter.com,2007:site.56945-856752</guid>
		<pubDate>Wed, 14 Feb 2007 08:17:11 -0800</pubDate>
		<dc:creator>blag</dc:creator>
	</item>
	</channel>
</rss>
