<!DOCTYPE HTML PUBLIC "-//W3C//DTD HTML 4.0 Transitional//EN">
<HTML><HEAD>
<META http-equiv=Content-Type content="text/html; charset=us-ascii">
<META content="MSHTML 6.00.2900.3132" name=GENERATOR></HEAD>
<BODY>
<DIV><FONT face=Arial size=2>
<DIV style="DIRECTION: ltr">Hi,<BR><BR>My name is Jean Quirion and I am a DSP
engineer. Currently I am<BR>working on a project where it is desired to
implement a VoIP solution<BR>over a GSM GPRS link. I would like to use Speex as
the vocoder for<BR>this application. This application would require the
Speex<BR>encoder/decoder and possibly the pre-processor to run on a low
power<BR>fixed-point DSP such as a TI C55x.<BR><BR>Thus, I am interested in
getting involved in the fixed-point<BR>implementation of Speex. I feel I can
contribute significantly to the<BR>Speex project while effectively developing a
vocoder solution for my<BR>project.<BR><BR>Can you tell me how I can get
started? I looked at the list of tasks<BR>on the Seepx wiki and I find that the
following would be best suited<BR>for my interests:<BR><BR>Complete fixed-point
(DSP development)<BR>• Rest of the narrowband modes<BR>•
Preprocessor (noise suppression, AGC)<BR>•
Jitter buffer<BR><BR>Please let me know how to move forward. Where does
the Speex vocoder<BR>fixed-point implementation stands? Has if been successfully
ported and<BR>tested onto a TI DSP evaluation board (i.e. an EVM)? If not, and
if it<BR>can be helpful, I would be interested in tackling that
task…<BR><BR>Sincerely,<BR></DIV>
<SCRIPT><!--
D(["mb","\u003cdiv style\u003d\"direction:ltr\"\>\u003cspan class\u003dsg\>\u003cbr /\>Jean Quirion\u003cbr /\>\u003c/span\>\u003c/div\>",0]
);
D(["ce"]);
//--></SCRIPT>
<DIV style="DIRECTION: ltr"><SPAN class=sg><BR>Jean
Quirion</SPAN></DIV></FONT></DIV></BODY></HTML>