[Opensim-dev] Mumble & Voice

[Matti Kuonanoja]

> I have an implementation of SLVoice.exe based on Mumble, running since a few
> weeks now. It supports speaker indication, but it runs for now only on
> Windows and does not yet support positional sound and private channels. The
> code still needs some cleanup and documentation about dependencies and the
> environment.
>
> Cheers, Volker
>

Nice job Volker!

Good to here you have got Mumble based SLVoice.exe working. I really
like to test it and to hear more about implementation details.

I think Mumble is definitely a technology to be considered to be used
for voice systems for virtual realities. It's open source with a
reasonable license. It also has a good audio quality and channel
system.

To use separate SLVoice.exe or not is clearly a client side design
issue and does not affect to the mechanism how client and server
negotiates about voice service capabilities.

What I like to see is a common agreement about the way how server can
inform clients about existing voice services. One possible way might
be a capability url where connections details can be obtained by the
client.

We have been using Mumble for in-world voice with Naali viewer and
Taiga server for a couple of months now. We are using region module to
offer mumble server connection details and client connects directly to
given Mumble server.

You can find more information about the implementation here:
http://wiki.realxtend.org/index.php/Mumble_integration

You can also try mumble voice yourself with Naali 0.2.5
(http://code.google.com/p/realxtend-naali/downloads/list) connecting
to world.realxtend.org:9000 server.


Cheers,

Matti Kuonanoja ~ realXtend developer