<span>H</span>i Haoyu, something like simple phrase recognition, or disctimination between pronunciation of a small list of options, can get you quite far in terms of verbal interfaces or flashcard/recitation practice.<br>
<br>I think perhaps some of the people interested in mentoring text to speech might also be interested in this. <br><br>SJ<br><br><div class="gmail_quote">On Tue, Apr 1, 2008 at 12:44 PM, Haoyu Bai <<a href="mailto:divinekid@gmail.com">divinekid@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;">Hello,<br>
<br>
I am a student from China. I have experience with speech recognition<br>
algorithms and practice. I am thinking add speech recognition ability to<br>
OLPC.<br>
<br>
I have read this wiki page: <a href="http://wiki.laptop.org/go/Speech_recognition" target="_blank">http://wiki.laptop.org/go/Speech_recognition</a><br>
<br>
In my opinion, since XO don't have a fast cpu, normal speech recognition<br>
algorithm and engine like Sphinx may not suitable. PocketSphinx may be<br>
faster, but its' training progress still slow.<br>
<br>
If we aim to some simple application like voice command, algorithm like<br>
DTW can be used, it is faster, but only can do simple phase recognition.<br>
<br>
I have no clear idea right now. I want to know is there a mentor<br>
interested in it? So we can discuss it deeply.<br>
<br>
Thank you!<br>
<br>
<br>
Best regards,<br>
<font color="#888888"><br>
Haoyu Bai<br>
4/2/2008<br>
_______________________________________________<br>
Gsoc mailing list<br>
<a href="mailto:Gsoc@lists.laptop.org">Gsoc@lists.laptop.org</a><br>
<a href="http://lists.laptop.org/listinfo/gsoc" target="_blank">http://lists.laptop.org/listinfo/gsoc</a><br>
</font></blockquote></div><br>