Hi,
Nice to see people asking for these specific features, especially when
we've a positive answer.
If you're on a linux box with alsa support, you can use input.alsa()
to get input directly from the microphone. Notice that it _always_
inputs from the microphone. If your normal stream is "s" then you can
create what you want using add([s,input.alsa()]). It will mix the two
streams together. If there is no sound on the mic, then you'll just
hear s renormalized by 1/2.
You can also use input.http() which relays a ogg/vorbis http stream,
typically streamed by an icecast server, to which your live show is
sent (using darkice for example). The latter solution seems more
complicated, but it may be more robust, doesn't require alsa on the
liquidsoap machine, doesn't require the dj to be on the liquidsoap
machine, and allows you to shut down the live show. Using input.http
you can do add([s,input.http("http://host:8000/live.ogg";)]).
Add as a weigths parameter with which you might want to play. If you
need more interactive control you might want to take a look at mix()
instead of add(), and consider using liguidsoap for controlling it.
But start with add().
Hope this helps. It should work, at least. If you need something more
specific, please describe precisely and I'll be happy to discuss how
to fine-tune your script.
Have fun.
--
David
