[theora] normalization of audio when using ffmpeg2theora?

[j at v2v.cc]

> We created some amateurish videos during Software Freedom Day. I like to
> publish them, but some are too quiet. How I can normalize the audio,
> when using ffmpeg2theora? Are you using some scripts to split audio and
> video and combine them back or what is the "automated way", how to
> achieve that? Enhancement of video will be useful too, but audio is
> critical.

Hi,
you have to split audio into an extra file, with that you could use
normalize-audio to adjust sound, something like this:

ffmpeg -i $input $input.wav
normalize-audio $input.wav
ffmpeg -i $input -i $input.wav -map 0.0:0.0 \
                               -map 1.0:0.1 \
       -vcodec copy -acodec copy \
       $input.normalized-audio.mkv
ffmpeg2theora $input.normalize-audio.mkv

ffmpeg2theora itself does not have any audio postprocessing, there are
some options for video check --pp help

j