Here are a couple of pointers to help you out with sync markers.
Sync markers are 00:00:00 based:
Even if you specify an offset timecode for your caption outputs, the sync markers need to be zero based.
Example: If you want to isolate the music in the audio 12 minutes after the beginning of the program until 19 minutes after the beginning of the program (even though your non-linear editor might tell you it's 01:12:00 to 01:19:00), you need to have markers like this:
Here's some music. ^M00:12:00 [ Music ] ^M00:19:00 Thanks Bob, that was great.
Sync markers are in true elapsed time (or drop frame timecode):
You cannot use non-drop frame timecode. You need to have your sync markers reflect the true elapsed time from the beginning of the program (drop frame is a close enough approximation). It is not relevant what frame rate the actual media has, since the marker timecode is expressed as a simple 30 fps value.
Complete documentation on transcription formatting and markers is available in our Transcription Guidelines article. A complete description of how to use sync markers in the transcript is available in our Sync Marker Summary article.