Sending a second emails since this is a second topic. I am working with PJSIP 2.2 and the C++ API. Apparently it seems that the C++ API is limiting the programmer to use a single audio device. I have introduced a new class in my test project which abstracts the pjsua's C API and is capable given the ID of a device to create a new instance of that device IFF is the devices has both inputs and outputs. During initilization I set the default AudioDevice of the Endpoint to null and I choose manually to which "Custom" Audio device I connect the media of a call instead of calling Endpoit::instance().audioManager().getDefaultPlaybackDev() (something like that) Is there any interest of integrating something like that in PJSIP? Best, Alex -------------- next part -------------- An HTML attachment was scrubbed... URL: <http://lists.pjsip.org/pipermail/pjsip_lists.pjsip.org/attachments/20140617/6d544c15/attachment.html>
