[Libav-user] Best File Container Format for editing Video/Audio?

James Board jpboard2 at yahoo.com
Wed May 14 19:48:45 CEST 2014

I wrote a libAV-based application which edits a video file.  It takes a video
file as input, along with the Edit List.  The input video file is an AVI container 
and was created by ffmpeg or mencoder during the inverse-telecine step.
The Edit list contains a list of Start/Stop PTS pairs.  The output is simply 
another video file based on those Start/Stop PTS pairs.

It's fairly simple.  But, it ignores audio and I want to modify it to include audio.
What can I do to make this easier to code in libAV?  Choose another container?

Also, in this thread, I might be using the words 'frame' and 'packet' incorrectly.

With AVI files, this seems complicated since the number of audio frames does not 
always equal the number of video frames.  So, I'd have to split the audio packets or 
frames into individual samples, and decide which samples should be in the output file.

Is there an easier way?  Is there a container format where the number of audio
packets or frames exactly equals the number of video packets?  If so, this could be a lot
easier for me.  Whenever I include a video frame, I would only have to include
the corresponding audio frame.  Does that make any sense?


More information about the Libav-user mailing list