Voip applications

My goel is to record the incoming voice in a voip application (which I don't
have the source). I know the codec being used (assuming I can use it in my
hooked code) and I plan to hook the win32 socket functions and extract the
RTP data. In the hooked code, I want also to build some graph that will
contain a source being fed with that data, a codec instance and some
renderer that will record to the disk.

Since I'm really new to this DirectShow amazing technology I'd like to hear
some opinions before I delve into the hard work.

Can someone in this group refer to that. Is it doable?

Thanks in advance