Jay Jarvinen wrote:
Perhaps http://aubio.piem.org/ or something similar, could be used to
map out a rough words-per-minute count.
Pitch detection is another idea, odds are that the voices/sounds in
commercials are dissimilar to the program being recorded. Perhaps
pitch-variation is a better term for what I'm suggesting.
There's also subtile/CC data, which I'd have thought would be a fair bit
more machine-readable, and could perhaps be bayesian filtered in much
the same way as email.
3-minute gaps in subs (or a sudden brief appearance of subs in an
unsubtitled programme) would be a really easy cue for the commercial
flagger to use.
kim.
_______________________________________________
mythtv-dev mailing list
[email protected]
http://mythtv.org/cgi-bin/mailman/listinfo/mythtv-dev
