removing laughter track

[victoriabears]

All I could find was this:-

http://cnx.org/content/m15641/latest/

but has anyone succeeded in removing this very very annoying aspect ?

[jagabo]

It's no surprise they didn't provide any before and after samples.

In the end, we found that detecting and removing laugh tracks from an audio signal is much more complicated than it may appear on the surface. As a human, it is very easy to spot a laugh in the signal, but to create a system that can do this automatically is more complicated. Another interesting aspect of the system is how to remove the laugh tracks, which can be complicated because people often talk over the laugh tracks in our signals. To remove only the laugh track while leaving in the human voice signals is something that needs further exploration.

In short, unless the laugh track is on its own track, you won't be able to remove it without causing severe damage to the rest of the audio.

If you have an AC3 source, you might be able to remove the left and right channels and leave only the center and maybe back channels.

[DB83]

It's like 'noise reduction' on audio. You set the scale to remove the noise but end up removing real sound on the same frequency.

[Lodovico]

iZotope RX's Spectral Repair filter may be able to tackle that. It's intended for removing things like coughs and chair squeaks in live music recordings, or the car squeal in this YouTube demo: http://www.youtube.com/watch?v=hbFe73nDNRU

It's a Pro tool with a price tag to match.

[jagabo]

Unfortunately, laughter will be in the same frequency range as the actors' voices. So spectral removal won't work well.

[zoobie]

ha ha ha

[AlanHK]

Trying to filter spectrally alone will give bad results, as it's human voice and it will kill the actor's voices as well.

However, the laughs are a reaction to something that has happened, or a spoken joke.
In a sitcom they normally pause after a joke to allow for audience reaction.
Usually there is no overlap with audience laughter and dialogue.

I doubt there is an automatic way of doing it, but if you have a great deal of patience, you could apply a very strong filter manually and selectively in Audacity, say.