<html><head><style type="text/css"><!-- DIV {margin:0px;} --></style></head><body><div style="font-family:courier,monaco,monospace,sans-serif;font-size:12pt">The proper approach for this would be to use the media_bugs which are <br>used to record the session have have been recently enhanced to displace<br>the actual audio with audio you can replace.<br><br>each streaming protocol can implement it's own mod<br>say "mod_shoutcast" and have an application or api command interface<br>hook to attach itself to the channel.<br><br>media bugs is a concept where you can latch a struct and a callback onto a channel. Your callback will be called continuously when audio is read or<br>written to a channel and you can choose on setup if you want it to be called<br>on read/write with muxed audio from both sides of a call or on write giving<br>you a chance to replace the audio.<br><br><span><a target="_blank"
href="http://www.freeswitch.org/docs/group__mb1.html">http://www.freeswitch.org/docs/group__mb1.html</a></span><br><div> <br>If you look in switch_ivr.c in switch_ivr_record_file<br><span><a target="_blank" href="http://fisheye.freeswitch.org/browse/FreeSWITCH/trunk/src/switch_ivr.c?r=4489">http://fisheye.freeswitch.org/browse/FreeSWITCH/trunk/src/switch_ivr.c?r=4489</a></span><br><br>You can see this is how it's possible to record from the background.<br><br>If you look at mod_ivrtest.c (basically a scratch pad to test code)<br><span><a target="_blank" href="http://fisheye.freeswitch.org/browse/FreeSWITCH/trunk/src/mod/applications/mod_ivrtest/mod_ivrtest.c?r=4400">http://fisheye.freeswitch.org/browse/FreeSWITCH/trunk/src/mod/applications/mod_ivrtest/mod_ivrtest.c?r=4400</a></span><br><br>the bugtest function shows how to set the channel up to run a bug<br>that gives you a chance to replace the frames as they are supposed to be <br>written to the channel so you could
attach your shoutcast socket and whenever<br>audio was to be written to the channel you could replace the frame with one<br>from your stream so the caller would hear the stream instead of silence<br>or whatever the channel was really writing etc.<br><br><br>When a bug is active on a channel it forces trancoding even when it's not <br>necessary for a bridge so you can always get the audio as raw SLIN which is <br>the only common audio format you can work with when combining media.<br><br><br><br><br><br><br></div><div>Anthony Minessale II<br><br><span>FreeSWITCH <a target="_blank" href="http://www.freeswitch.org/">http://www.freeswitch.org/</a></span><br><span>ClueCon <a target="_blank" href="http://www.cluecon.com/">http://www.cluecon.com/</a></span><br><br>AIM: anthm<br>MSN:anthony_minessale@hotmail.com<br>JABBER:anthony.minessale@gmail.com<br>IRC: irc.freenode.net #freeswitch</div><div><br>FreeSWITCH Developer
Conference<br>sip:888@conference.freeswitch.org<br>iax:guest@conference.freeswitch.org/888<br>googletalk:conf+888@conference.freeswitch.org<br>pstn:213-799-1400</div><div style="font-family: courier,monaco,monospace,sans-serif; font-size: 12pt;"><br><br><div style="font-family: times new roman,new york,times,serif; font-size: 12pt;">----- Original Message ----<br>From: Mark D. Anderson <mda@discerning.com><br>To: freeswitch-dev@lists.freeswitch.org<br>Sent: Thursday, March 8, 2007 5:04:50 PM<br>Subject: [Freeswitch-dev] adding support for incoming streaming audio<br><br><div>I'd like to have a streaming audio source (not just playing a local file)<br>for users while they are parked or waiting for a conference to start.<br><br>In digging into FS it seems there are several ways this<br>might be approached. Of course this is based on only a superficial<br>understanding of the FS internals.<br>But I've come up with 6 (count them, 6) possibilities:<br><br>1) hack the
"playback" application.<br>Right now, the lookup of driver in switch-core.c is by file suffix<br>(with no override in xml conf).<br>Currently the only real format module is mod_sndfile.<br>(The string constants for metadata like SWITCH_AUDIO_COL_STR_TITLE <br>are conveniently set to be identical to the enum used by libsndfile.)<br><br>It could be done by hacking mod_sndfile, or instead by<br>making a new format driver, say "mod_url".<br>libsndfile does not support mp3 (and won't), and is oriented around local files.<br>So a mod_url driver would be better, with support for some suffixes <br>like .pipe or .m3u or .mp3. It would need to do content-type sniffing<br>in some cases.<br><br>One issue is that it seems that mod_playback is blocking til completion.<br>There is a timer in switch_ivr_play_file; it would be necessary to invert<br>control there to allow the processing to proceed without the file ending.<br><br>Pros: most useful; wherever a local
file is used, a remote could be used instead.<br>Cons: non-trivial amount of hacking.<br><br>2) Use an external process to convert streaming audio into sip.<br>For example, using mjua or pjsua as sip clients, along with some<br>command-line streaming audio client such as streamripper.<br><br>Pros: no change to FS required<br>Cons: lots of moving parts, and the SIP gateway utility will probably require some hacking.<br><br>3) implement a "mod_shout" as an endpoint for shoutcast/icecast protocol<br><br>Pros: fits in well with FS architecture<br>Cons: requires understanding FS internals concerning threading, buffering, etc.<br><br>4) implement a "mod_rtsp" as an endpoint.<br>This would mean I'd have to be satisfied with supporting only<br>rtsp streaming audio, not shoutcast/icecast streaming audio.<br><br>Pros: In theory there wouldn't be much to do, since there is already RTSP<br> support in the FS stack.<br>Cons: requires underestanding FS
internals *and* RTSP streaming audio<br><br>5) hack mod_conference to do stream-in just like record-out, as a thread.<br>[I'm not sure why mod_conference does it that way, rather than at the core<br>switch level (which *also* has session recording ability) -- isn't the point<br>of a softswitch framework that it can allow<br>for arbitrary combinations of modules? Shouldn't there be a way to<br>supply an arbitrary channel as a member?]<br><br>Pros: relatively easy to do by blindly copying the session record case.<br>Cons: specific to conferencing, and makes mod_conference even bulkier.<br><br>6) implement mod_park with specific support for streaming audio<br>[right now mod_park is just an empty stub.]<br><br>Pros: standalone so easier to implement<br>Cons: limits its usefulness to just that "application"<br><br>_______________________________________________<br>Freeswitch-dev mailing list<br>Freeswitch-dev@lists.freeswitch.org<br><a target="_blank"
href="http://lists.freeswitch.org/mailman/listinfo/freeswitch-dev">http://lists.freeswitch.org/mailman/listinfo/freeswitch-dev</a><br>UNSUBSCRIBE:<a target="_blank" href="http://lists.freeswitch.org/mailman/options/freeswitch-dev">http://lists.freeswitch.org/mailman/options/freeswitch-dev</a><br><a target="_blank" href="http://www.freeswitch.org">http://www.freeswitch.org</a><br></div></div><br></div></div><br>
<hr size=1>Looking for earth-friendly autos? <br> <a href="http://autos.yahoo.com/green_center/;_ylc=X3oDMTE4MGw4Z2hlBF9TAzk3MTA3MDc2BHNlYwNtYWlsdGFncwRzbGsDZ3JlZW5jZW50ZXI-">Browse Top Cars by "Green Rating"</a> at Yahoo! Autos' Green Center. </body></html>