Double Caption Data in ATSC

Decoding ATSC broadcast with "mpegvideoparse ! avdec_mpeg2video" results in double caption bytes entering the pipeline.

Both elements insert decoded caption data, resulting in duplicate bytes. Unsure of the caption data order in relation to picture frame order when using mpegvideoparse versus avdec_mpeg2video.

Example pipeline:

/opt/gstreamer/bin/gst-launch-1.0
dvbsrc frequency=531000000 modulation="QAM 256" delsys=atsc ! queue !
tsparse ! tsdemux program-number=1006 name=demux
demux. ! queue ! video/mpeg ! mpegvideoparse ! avdec_mpeg2video ! tee name=dvideo ! videobin.
demux. ! queue ! audio/x-ac3 ! ac3parse ! avdec_ac3 ! queue ! audioconvert ! audio/x-raw,rate=48000,format=F32LE,channels=2 !
audioresample ! pulsesink
dvideo. ! queue ! video/x-raw ! ccextractor ! closedcaption/x-cea-708,format=cc_data ! tee name=cdp !
queue ! cc708overlay name=videobin font-desc="DejaVu Sans 12" !
textoverlay text="WTTG-HD" halignment=right valignment=top shaded-background=true ! queue ! video/x-raw ! autovideosink
cdp. ! queue ! zmqsink endpoint=tcp://*:5557 bind=true