TechOps Processing of Videos includes CC Embedding, Subtitling, Open Description, Expanded Description, and SAP proxies. However, clients primarily send large original videos for CC Embedding and Subtitling; the other products aren't as standardized and we can use smaller videos for them.

TechOps' default practice is to match the source video and audio as closely as possible. Whatever the client exported and sent to us is what they'll get back (along with the specific accessibility components they ordered of course), unless we are directed otherwise. So if a client needs a very specific video/audio set-up that isn't already in the video they sent to us, we will need to know those specific specs. Usually, if the client never mentions what kind of video/audio specs they want, matching the provided source is fine for them. 

That said, for embedding or rendering we’d normally expect/want an MXF or ProRes MOV with a minimum of 720 x 480 resolution, or at least something that has 16x9 resolution. If the video we use has a low resolution the subtitles won’t look good when the video is enlarged. We want their highest quality video since we’ll be exporting it with those same specs when we send it back to them.

Ideally we'd have the client send a smaller proxy alongside the large original for embedding/rendering. But if they can't/don't want to that, then TechOps can create a smaller proxy for production from the large original.

If the client has an mp4 and wants captions embedded in it, that’s a red flag. Mp4s are not very high quality and a lot of players for them don’t support captions. Subtitles will probably serve the client better (although we will still embed an mp4 if the client insists).