[mythtv-users] Re: Automatic Commercial Detection - how to get it

Gavin Hurlbut gjhurlbu+mythtv-users at beirdo.ca
Wed Sep 22 10:36:14 EDT 2004

Chris Pinkham <cpinkham at bc2va.org> wrote:
> I've yet to look at the audio data enough to know this or not, but I've heard
> and read that it's not volume, it's something else that changes that makes
> it seem louder.  I don't know.  I haven't even delved into trying to process
> the audio to detect commercials, that's a whole 'nother beast.  Things like
> audio silence during blank-frames and such would really help proper
> detection.

I think that the commercials are more compressed (using dynamic compression,
not like MP3 compression), so they appear louder to the human ear.  Also, they
may also be boosting the levels after the compression.  I remember about 10
years ago listening to a Blue Jays game on the radio, and all the commercials
cut out to silence because they boosted the gain on the compressed commercials
just a bit too far, and the satellite uplink dropped the clipped audio.  It
made for a more pleasant game.

Detecting audio silence during blank-frames may well prove to be a useful 
addition.  I'm not sure how much more CPU overhead that would cause, and you
may need to use a DSP algorithm to get it "foolproof", but it would likely
help some.

That being said, what we have now is pretty accurate for me.  A few shows like
CSI and Law & Order quite regularly fooling it, but that's life.  What do we
really expect for free software done in people's spare time :)   If we want
to make it better, nobody's stopping us from trying.


More information about the mythtv-users mailing list