Just a question on how Wowza handles this. I have a flash player which allows publishing of webcam (in Sorenson Spark) to the Wowza server. This in turn can be requested using rtsp://. When I look at the SDP file the stream is being split into two rtp streams.
Audio: SPEEX
Video: H263-1998
Two questions:
-
Is there a way within the Wowza API to adjust the SDP files that are generated so that I can correctly identify the video stream as Spark instead of H.263-1998?
-
What is the encapsulation of the spark video in the RTP stream? I have traced it with wireshark it does not appear to be either H.263 over RTP or H.263-1998 over RTP format. I am trying to utilize the ffmpeg FLV1 decoder on the stream but it is producing mostly garbage.
Any help would be greatly appreciated. Thanks,
Ehren