Re Audiobooks: Synchronization of audiobooks with ebooks would be highly desired. There are archives and archives of human read audiobooks provided by <a href="http://Librivox.org">Librivox.org</a> that are currently being transcoded into the Highly Compressed speex format. With Voice optimized speex it is possible to encode an hour of audiobook in about 2-4 mb.
<br><br>At this point it would probably be easier to implement some sort of highlighting program and tagging in the speex file. As the audiobook hits specific time frames it highlights the sentence that is being read. Word by word tagging could be used for really early literacy materials.
<br><br>I've been meaning to write up a draft and use-case for this program. But I've been working on packaging up the audiobooks still. I'll go draft something up on the wiki tonight.<br><br>Seth<br><br><div class="gmail_quote">
On Dec 3, 2007 12:51 AM, Charbax <<a href="mailto:charbax@charbax.com">charbax@charbax.com</a>> wrote:<br><blockquote class="gmail_quote" style="border-left: 1px solid rgb(204, 204, 204); margin: 0pt 0pt 0pt 0.8ex; padding-left: 1ex;">
Yeah, or how about synchronizing of an audiobook with an ebook, where the synchronisation was done with speech recognition on a server which outputs a synchronisation file with the audiobook file so that it can provide this karaoke style boldening of the text or the line of text that is being read on the audiobook.. Even though 9-10 hour audiobooks are over 100mb at a decent quality encoding, there is enough space on the XO for one or two of them at a time, and they could be downloaded to the school server overnight or cheaply through a satellite download connection.. Then also have a 400mb or so video version of the story with subtitles..
<div><div></div><div class="Wj3C7c"><br><br><div class="gmail_quote">On Dec 3, 2007 2:34 AM, Edward Cherlin <<a href="mailto:echerlin@gmail.com" target="_blank">echerlin@gmail.com</a>> wrote:<br><blockquote class="gmail_quote" style="border-left: 1px solid rgb(204, 204, 204); margin: 0pt 0pt 0pt 0.8ex; padding-left: 1ex;">
I assume<br>that the computer will read to children and parents alike when<br>multilingual text-to-speech software is included. Karaoke-style<br>coloring of the text is also important to show what letter or syllable<br>is being pronounced. We know that children will learn to read if they
<br>hear and see stories at the same time. Some unknown fraction of the<br>parents will also learn to read under those conditions, just is<br>illiterate Indians learn to read while singing along with captioned<br>Bollywood movies.
<br></blockquote></div><br clear="all"><br></div></div><font color="#888888">-- <br>Charbax,<br>Nicolas Charbonnier
</font><br>_______________________________________________<br>Olpc-open mailing list<br><a href="mailto:Olpc-open@lists.laptop.org">Olpc-open@lists.laptop.org</a><br><a href="http://lists.laptop.org/listinfo/olpc-open" target="_blank">
http://lists.laptop.org/listinfo/olpc-open</a><br><br></blockquote></div><br>