Re: [codec] #3: 2.2. Conferencing: Support of binaural audio?

[Stefan Sayer <[e-mail]>]

o Gregory Maxwell [03/23/2010 07:59 PM]:
> Christian Hoene [[e-mail]] wrote:
>> (a) does not have any impact on the requirements but in case of (b) codec requirements are the support of stereo speech transmission and support for efficient mixing.
> 
> I agree that (b) requires support for stereo. 
I, too, agree on that, and I think stereo encoding support is required, 
and for spatial audio conferencing with centralized mixing it is 
sufficient. The question here would be: Can the codec optimize stereo 
encoding, if it is known to the codec that the stereo signal contains 
sources which are spatially rendered with known angles (because the two 
signals are strongly correlated with a known time difference)? Or will 
the usual methods employed in stereo encoding already catch this?

It is also my view that for decentralized mixing, a positioning 
mechanism should not be part of the codec.

Note also that support for stereo and multi-channel is also explicitly 
mentioned in applications 2.3 Telepresence and 2.6  Live distributed 
music performances / Internet music lessons.

> Of course, minimizing the codec computational burden is helpful for scaling conferencing systems.
definitely

> 
> Can anyone share some typical conference scaling numbers which they consider interesting?    
 > It's been my view that all comers are fast enough that conferencing 
on commercial scale isn't
 > much of an issue:  That, yes, it might take some serious processing 
grunt to handle 10,000 users on a conferencing system,
 > [...]

for PSTN quality (G.711) conferencing, today a good PC based server 
without any extra DSP hardware can handle 5000 users, but that drops to 
about 60% of it if you use GSM codec, and about 23% for iLBC or speex 
codec. so, minimizing the computational burden is indeed helpful.

Regards
Stefan