Format of an SDP file for audio - AWS Elemental Live

Format of an SDP file for audio

v=0 o=- 1443716955 1443716955 IN IP4 10.xx.xxx.236 s=st2110 0-1-0 t=0 0 m=audio 20000 RTP/AVP 97 c=IN IP4 239.x.x.x/64 a=source-filter: incl IN IP4 239.x.x.x 10.xx.xxx.236 a=rtpmap:97 L24/48000/2 a=mediaclk:direct=0 rate=48000 a=framecount:48 a=ptime:1 a=ts-refclk:ptp=IEEE1588-2008:04-5c-6c-ff-fe-0a-53-70:127

Following is information about the data in this example:

  • o= This line identifies the source IP for the stream

  • m=audio. This identifies the file as an audio description. 20000 is the port of the stream. This line must occur before the a lines.

  • c= identifies the destination IP address. This is a unicast or multicast address.

  • a=source-filter identifies a filter. This line is optional.

    In a file for an input to AWS Elemental Live:

    • If the line is included, AWS Elemental Live will listen only to the source IP (10.xx.xxx.236) for packets.

    • If the line isn't included, AWS Elemental Live will listen to any packets on the destination IP address (239.x.x.x).

    In a file for an output: AWS Elemental Live always includes this line in any SDP file that it creates.

  • a=rtpmap provides information about the audio format:

    • An example for PCM:

      a=rtpmap:97 L24/48000/2

      L24 is the number of bits per PCM audio sample. 48000 is the sample rate. 2 is the number of channels, in this example, a single stereo track.

    • An example for a Dolby Digital codec:

      a=rtpmap:96 AM824/48000/5 a=fmtp:96 channel-order=SMPTE2110.(AES3,AES3)

      48000 is the sample rate. 6 is the number of AM824 subframe pairs, and must be an even number.