RTPInsertSound tool to insert audio into a specified audio (i.e. RTP) stream was created in the August – September 2006 timeframe. The tool is named rtpinsertsound. It was tested on a Linux Red Hat Fedora Core 4 platform (Pentium IV, 2.5 GHz), but it is expected this tool will successfully build and execute on a variety of Linux distributions.
-a source RTP IPv4 addr -A source RTP port -b destination RTP IPv4 addr -B destination RTP port -f spoof factor - amount by which to: a) increment the RTP hdr sequence number obtained from the ith legitimate packet to produce the RTP hdr sequence number for the ith spoofed packet b) multiply the RTP payload length and add that product to the RTP hdr timestamp obtained from the ith legitimate packet to produce the RTP hdr timestamp for the ith spoofed packet c) increment the IP hdr ID number obtained from the ith legitimate packet to produce the IP hdr ID number for the ith spoofed packet [ range: +/- 1000, default: 2 ] -i interface (e.g. eth0) -j jitter factor - the reception of a legitimate RTP packet in the target audio stream enables the output of the next spoofed packet. This factor determines when that spoofed packet is actually transmitted. The factor relates how close to the next legitimate packet you'd actually like the enabled spoofed packet to be transmitted. For example, -j 10 means 10% of the codec's transmission interval. If the transmission interval = 20,000 usec (i.e. G.711), then delay the output of the spoofed RTP packet until the time-of-day is within 2000 usec (i.e. 10%) of the time the next legitimate RTP packet is expected. In other words, delay 100% minus the jitter factor, or 18,000 usec in this example. The smaller the jitter factor, the greater the risk you run of not outputting the current spoofed packet before the next legitimate RTP packet is received. Therefore, a factor > 10 is advised. [ range: 0 - 80, default: 80 = output spoof ASAP ] -p seconds to pause between setup and injection -h help - print this usage RTPInsertSound -v verbose output mode RTPInsertSound
cyborg@cyborg:~$ sudo rtpinsertsound a.wav -v Targeting interface eth0 libfindrtp_find_rtp(): using pcap filter "ip".