Audio editing of mpg files-expert advice needed

I have some mpg files that are interviews of people at a fair. I've been asked by the person that shot the videos for a college project to try to reduce the background sounds and bring up the vocals. Unfortunately, he did not use a remote mike so all sound being picked up comes from the mike on the camcorder. Sound quality is decent and the mpg image quality is very high (i.e. very little if any compression used in the transfer from camcorder to mpg). Background sound level is pretty high though and not consistent. S/N ratio maybe 2:1 at best. Think cows, pigs and sheep here.

My first inclination is to try to open the files in SoundForge 6 or VegasVideo and just try equalizing the sound with the graphic equalizer and then saving the resulting file. I have decent experience in audio editing of wav files, but with consistent background noise such as tape hiss. That I can handle. I have very little experience editing audio that is part of a video file.

I'm open to any ideas such as what editing software to use, any plug-ins that might be recommended for increasing the level of spoken word, and even what plugins I'll have to have to be able open the mpg files.

I have many audio and video editing programs and plugins I've collected during the "bad old days." Now its time to turn around and start using them.

Thanks
 
you may find the techniques here useful
h**p://www.a-reny.com/iexplorer/restauration.html

With a profiling noise reduction, ytou may have to re-profile and work on smaller segments, if you can form a reasonably effective local noise profile for each - if you don't have a suitable tool from "the old days", the wave editor in EAC has profile-based noise reduction, and proved potent in a PC fan noise test.


Err - do you mean the Cows, pigs and sheep ARE the background noise?? - I thought that was called atmosphere!

One "quick and dirty" suggestion is to use anything which can save it with GSM compression - as GSM has considerable suppression of non-voice sounds.
 
Last edited:
I've never heard of GSM compression. Please elaborate. Most of the background noise is the chatter of other people talking like crowd noise.
 
It's what mobile phones use!

Mono only, but varying sample rates - it's available to Windows Sound Recorder, not sure what else.

As I said, it's a "quick and dirty" that MIGHT emphasize the strongest voice at the expense of everything else, since as I understand it, GSM effectively regenerates the voice sounds by rule.
 

nb

1
There are a number of plugins but if I'm not mistaken SF comes with it's own compressor that will increase the vocals as one of its settings then using another increase the overall volume of the stream
 

nb

1
Just look for the name compressor/compression or multiband compression and just play with it/them.
 
Top