LIVE audio stereo stream

Question

Hi there,
I'm looking for a way to offer a LIVE audio stereo stream to
the web client using Shockwave plug-in.
I was "banging my head off forums and search engines" (as
flashcomguru nice says) for some weeks
to find a way to stream LIVE audio stereo to Flash Media
Server but I couldn't find one viable direct
solution for this platform.
The client Flash Media Server app uses Microphone.get method
to capture sound from soundboard
which unfortunately is mono. it uses a speech Nellymoser
codec, at which music sounds pretty OK at 44Khz.
Possible solution: I can capture the LIVE line-in stereo
sound using two soundcards and thus beeing able to
create two sepparate mono audio streams one for left, one for
right using the only available Microphone.get method.
It will then maybe need some synchronization server scripts
later.
The trick is: In Director I use two .swf movies (action
script controlled) to connect to server
and play the streams but I couldn't find a way to direct the
streams to a specific channel
(one of the 8 available) from which point I can use the LINGO
Sound Channel PAN property, like this:
sound(1).pan = 100 for right
sound(2).pan = -100 for left
From all that I read I came to the conclusion that you can
use the Sound Channel PAN property only to
.swa imported and prerecorded audio cast members, not to the
live audio stream which is played roughly
in the main sound output channel to which I can't apply the
Sound Channel PAN property.
The key is to route those two streams left and right in
Director.
Any hints?
Thanks for reading,
My deepest respect,
hOver

Answer 1

The microphone code is very similar to what you have posted.  I can successfully use the enhanced microphone.  When it is enabled, the issue I am having is exhibited.
A simple test I am using:
Use ffmpeg to stream a stereo mp3 file to the media server.  I am using the following ffmpeg command line:
ffmpeg -re -i ~/alone.mp3 -vn -acodec copy -ac 2 -f flv rtmp://myserver.com:1935/video/3Daudio/alone
In this case the file is encoded at 44.1 kHz.
The client uses a netstream to play with bufferTime = 0
Without the microphone, the playback is as expected.  With a normal microphone, not the enhanced microphone, the playback is as expected but there is little to no echo cancellation.
When the enhanced microphone is enabled, again using similar code to your post, the mp3 playback becomes severely distorted and is unacceptable.
In my opinion, this is an issue with the AEC algorithms of the enhancedMicrophone and stereo playback of a 'live' stream.  If I modify the client playback code to bufferTime > 0, the mp3 playback is normal but there is no echo cancellation.
Thanks,
Sam