Reputation: 13781
I'm writing an application that can "watch" a TV channel, interpret the closed captioning, and record the channel content to disk. I've never done any sort of programmatic interaction with video or closed captioning, so even basic advice could prove helpful.
I'm flexible on the technologies and the content sources (e.g. DirecTV, slingbox, tuner card, etc.).
Does anyone know of any toolkits or libraries which could serve as a starting point?
Upvotes: 0
Views: 1332
Reputation: 11
Closed Captions are stored in the User Data part of an MPEG2 TS Stream, or SEI part of an H264 stream, in any case you need to parse video access units, or NAL Units, and then once you extract a closed caption, you parse it to generate text. This last part might be doable using CCExtractor, however you need to decode video
Upvotes: 1
Reputation: 797
Take a look at CCExtractor. It might be a good starting point for what you need to do. Remember that MPEG compression will often ruin the video's embedded captioning data, so you'll need to be careful about where you're getting your source material from:
http://ccextractor.sourceforge.net/
Upvotes: 1
Reputation: 16934
I've personally never worked with "non-digit" content, but I'd start with looking into MythTV. It's an open source project for a DVR system.
Upvotes: 0