Basically, there's no optimal solution to the problem. Speech codecs
just don't handle mixed streams optimally. So you'll have to do tests
and decide which solution gives you the best results.
Jean-Marc
Le lun 10/11/2003 à 13:59, Oded a écrit :> I have received your answer regarding the intention of developing
> audio mix algorithm based on Speex (a while ago...).
>
> You've mentioned two different scenarios.
> VOIP applications that can manage with high bit-rates can truly mix
> the audio just once at
> the receiving client. Today most sound cards can mix the decoded
> streams automatically (no need for mix software).
>
> The real problem is decoding mixing and than rencoding (where bit-rate
> count...).
> From my experience as a VOIP (and video conference) appication
> developer I still think that mixing few audio streams in software is a
> major problem trying to keep audio quality...
>
--
Jean-Marc Valin, M.Sc.A., ing. jr.
LABORIUS (http://www.gel.usherb.ca/laborius)
Université de Sherbrooke, Québec, Canada
-------------- next part --------------
A non-text attachment was scrubbed...
Name: signature.asc
Type: application/pgp-signature
Size: 190 bytes
Desc: Ceci est une partie de message numériquement signée
Url :
http://lists.xiph.org/pipermail/speex-dev/attachments/20031110/f244e5d5/signature-0001.pgp