Hello Mihaly, Mihaly Zachar wrote: with SEMS you can easily build an application that will decode the RTP, save as wav file, and encode to the other call leg, possibly transcoding the call. An example for such a type of application is 'jukecall': http://ftp.iptel.org/pub/sems/doc/current/ModuleDoc_jukecall.html you'd only need to remove the ivr functionality, and tweak it a little so that audio is both relayed and saved to file. if on the other hand you need to do (il)legal interception type of application that will relay and save the RTP without doing anything with it, I think its simpler done with modified rtpproxy; afair such functionality is already in part implemented, or at least easy to implement. the difference is that SEMS has two independent calls (dialogs, sessions) for each leg; both legs are negotiated independently and only the (decoded) audio is connected between the two, and SEMS does the whole decoding, buffering, PLC etc. Alternatively you could also modify the SEMS B2BUA to fix the SDP and relay RTP, but for this type of application I'd say it's probably easier to use rtpproxy. for further SEMS related tips and implementation hints I'd suggest to move the discussion to semsdev mailing list (http://lists.iptel.org/mailman/listinfo/semsdev). hth Stefan -- Stefan Sayer VoIP Services stefan.sayer(a)iptego.com www.iptego.com iptego GmbH Am Borsigturm 40 13507 Berlin Germany Amtsgericht Charlottenburg, HRB 101010 Geschaeftsfuehrer: Alexander Hoffmann