Re: [codec] WG Review: Internet Wideband Audio Codec (codec)

Hi,

I also would like to add, that with Internet you can't control every participant
and require that he implements some extensions feature. This effectively
means that you can't rely on any external features, so having this ability
inside of a codec (which is an essential part and may be enforced) is very
much desired.

On Thu, Jan 14, 2010 at 04:15, Daniel Petrie <dpetrie@sipez.com> wrote:
> Hi Roni:
> I agree that transporting this information is code independent information.  However calculating these as part of the encoding may be possible significant performance savings depending upon the codec algorithm.
>
> Cheers,
> Dan
>
> --- On Wed, 1/13/10, Roni Even <ron.even.tlv@gmail.com> wrote:
>
>> From: Roni Even <ron.even.tlv@gmail.com>
>> Subject: RE: [codec] WG Review: Internet Wideband Audio Codec (codec)
>> To: dpetrie@sipez.com, "'Alexander Chemeris'" <Alexander.Chemeris@sipez.com>, "'Christian Hoene'" <hoene@uni-tuebingen.de>
>> Cc: codec@ietf.org
>> Date: Wednesday, January 13, 2010, 3:53 PM
>> Daniel,
>> These are not a codec specific information and there is
>> work on it in AVT.
>> See
>> http://tools.ietf.org/id/draft-lennox-avt-rtp-audio-level-exthdr-01.txt
>> and
>> there is also
>> http://tools.ietf.org/id/draft-ivov-avt-slic-02.txt
>> which provides
>> information to the receiver.
>>
>> Roni Even
>>
>> > -----Original Message-----
>> > From: codec-bounces@ietf.org
>> [mailto:codec-bounces@ietf.org]
>> On Behalf
>> > Of Daniel Petrie
>> > Sent: Wednesday, January 13, 2010 7:54 PM
>> > To: 'Alexander Chemeris'; Christian Hoene
>> > Cc: codec@ietf.org
>> > Subject: Re: [codec] WG Review: Internet Wideband
>> Audio Codec (codec)
>> >
>> > While we are discussing conferencing, I would like to
>> add my wish list
>> > of requirements:
>> >
>> > I would want the following information included in the
>> RTP packet so
>> > that I can do speaker selection on the streams without
>> having to have
>> > the expense of decoding them and run the algorithms to
>> calculate these
>> > values.  This is useful in all 3 of Christian's
>> conferencing scenarios.
>> >
>> > 1) Voice activity indication (minimally binary,
>> ideally magnitude)
>> > 2) gain control either already applied or a multiplier
>> to apply to
>> > achieve a consistent gain.
>> > 3) Energy level
>> >
>> > --- On Wed, 1/13/10, Christian Hoene <hoene@uni-tuebingen.de>
>> wrote:
>> >
>> > > From: Christian Hoene <hoene@uni-tuebingen.de>
>> > > Subject: Re: [codec] WG Review: Internet Wideband
>> Audio Codec (codec)
>> > > To: "'Alexander Chemeris'" <Alexander.Chemeris@sipez.com>
>> > > Cc: codec@ietf.org
>> > > Date: Wednesday, January 13, 2010, 5:55 AM
>> > > > This codec should be also useful
>> > > in
>> > > > conference
>> > > > scenarios, where you have to decode it at
>> conference
>> > > server and encode
>> > > > mixed signal again. And if I haven't missed
>> something,
>> > > this use-case
>> > > > was
>> > > > explicitly listed in codec requirements.
>> > >
>> > > The conferencing scenario is not yet well
>> described in the
>> > > current codec requirements document. Different
>> kinds of
>> > > systems are possible:
>> > >
>> > > 1) A central conferencing server that serves all
>> clients.
>> > > - a) which does a efficient single changing
>> mixing - still
>> > > a challenge if you have multiple operational
>> modes in a
>> > > codec.
>> > > - b) which does a more intelligent mixing using
>> (at least)
>> > > two channels to support spatial hearing - this
>> increases the
>> > > quality of conference call significant. Thus,
>> stereo modes
>> > > must be supported if the codec shall support
>> state of the
>> > > art conferencing technologies.
>> > > 2) A mesh-like N-to-N transmission conference
>> call. Then
>> > > you might benefit from layered coding because you
>> have at a
>> > > client you have to encode only once. But then
>> again, how to
>> > > support multiple coding modes?
>> > > 3) One client acts as conference server (similar
>> to 1)
>> > >
>> > > In total, the support of conferencing calls will
>> require:
>> > > - Layered coding
>> > > - Efficient mixing of streams
>> > > - Support of stereo
>> > > - Good (self-)transcoding
>> > > Is this all achievable?
>> > >
>> > > Christian
>> > >
>> > >
>> > > _______________________________________________
>> > > codec mailing list
>> > > codec@ietf.org
>> > > https://www.ietf.org/mailman/listinfo/codec
>> > >
>> > _______________________________________________
>> > codec mailing list
>> > codec@ietf.org
>> > https://www.ietf.org/mailman/listinfo/codec
>>
>>
>

-- 
Regards,
Alexander Chemeris.