Hello Jerome, the team that was running the CLI recognized they weren't running the fullscan correctly w/ ParseSpeed=1 and after resolving that the 608 captions are being detected. So false alarm, sorry about that. Closing the ticket.
wondering if it would be helpful for us to push a piece of sample media up for you to take a peek at?
thanks for your fast reply Jerome! Here's a response from our support team: Looking at the Starz Mezzanines and they all have .MPG file extensions There is content in the stream, we are able to extract a valid srt with CCextractor Also I do not think scan duration is at play here because 1) We ran full scan option from CLI 2) We have working assets whose captions fire much later in the stream than the assets that are failing caption detection
Captions are not detected
HDS encoding creates f4m manifest that can have segments/fragments in either a .f4x/.f4f...