Hi everyone,

I am writing a voip application using telepathy and I have run into some 
strange issues that I can't explain myself. Let me explain the situation. I am 
doing some testing with audio calls over jabber/jingle between my desktop 
computer and my laptop (both inside the local NAT, in case this matters).

First, I have tried doing calls between my client (on the desktop) and google 
talk (on the laptop). This scenario mostly works, I get correct audio on both 
sides, but after a while (~30 seconds or so) the audio stream from my client 
stops being delivered to gtalk and instead I get a continuous noise (always 
the same) on gtalk.

Second, I have tried doing calls between my client and empathy. In this case, 
although they connect fine, I don't get any audio on either side, instead I get 
a "tak tak tak" noise with exactly 1 Hz frequency. This noise seems to be 
coming from FsConference itself. I'll try to explain what makes me believe 
that: I have a gstreamer pipeline that looks like this:

audiosrc -> volume -> FsConference -> liveadder -> audioresample -> volume -> 
audiosink

(audiosrc and audiosink can be anything. I have tried so far autoaudio{src, 
sink}, alsa{src,sink}, oss{src, sink} and audiotestsrc.)

Empathy has a similar one that looks like this:

gconfaudiosrc -> volume -> level -> FsConference -> liveadder -> audioresample 
-> volume -> gconfaudiosink

The only practical difference is the "level" element, which apparently feeds a 
small visualization widget on the empathy window that shows the level of the 
audio coming from the microphone.

Both sides receive this "tak" noise. Now, if I mute one side by setting the 
volume on the "volume" element of the input to 0, the other side stops getting 
the noise. However, increasing the volume doesn't make any significant 
difference. An important detail is that the visualization widget on empathy 
reacts to the microphone and is not synchronized with the noise that I get on 
the other side. So, that leads me to the conclusion that the FsConference 
element somehow gets correct audio from the input and transforms it into this 
noise. Thus, if it gets no audio, there is no noise.

The same exactly happens when I am trying to make calls with my client on both 
sides. However, calls with empathy on both sides work fine. This makes me 
wonder what can be wrong with my client. Both my client and empathy are very 
similar applications, both using telepathy, telepathy-farsight, gstreamer and 
even (almost) the same pipeline.

Does anybody have any clue? What can be wrong?
Thanks in advance.

Best regards,
George

PS: I am using telepathy-farsight 0.0.7 and farsight2 0.0.12 from debian 
packages.


------------------------------------------------------------------------------
Enter the BlackBerry Developer Challenge  
This is your chance to win up to $100,000 in prizes! For a limited time, 
vendors submitting new applications to BlackBerry App World(TM) will have
the opportunity to enter the BlackBerry Developer Challenge. See full prize  
details at: http://p.sf.net/sfu/Challenge
_______________________________________________
Farsight-devel mailing list
Farsight-devel@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/farsight-devel

Reply via email to