how to separate dvd audio tracks?

J

JPWDW

Member
Hi everyone!

I have this instrumental track about Disney movies that someone recorded having their English character voices from the movies. I used Audacity and the tool it includes to remove part and most of the vocals from the track. Right now the track when you hear it sounds a little bad, but my guess is that once I can handle to put the Spanish character voices it will be heard better. I also have the movies, ie: Peter Pan, on Special Edition and there is an option where you hear only the back audio, like the soundtrack or score, and another on which you can hear the back audio plus the Spanish/Language voices included. Is there any way using any software or program or the like that I can separate those two tracks and take only the voices so then I can add them later to my instrumental track? Thanks in advance!
 
As far as I'm aware, you can only do that if the music and vocals are separate and on different channels. If they are not, there really isn't a way to split them apart as they are already mixed . I've never been able to find anything that does a decent job anyway.
I suppose it will depend if the DVD has the things mixed or not . I suspect that it's not a case of them adding the audio and voices separately but having a track with them mixed together for each language. have you tried something like DVDdecrypter or shrink to see if you can isolate and rip the individual audios on the DVD?
If there is way, I'd be interested to know from our audio experts here (which I'm not).
 
Thanks! I was able to accomplish this. I used a program called Smart Ripper and told it to encode the audio as a AC3 external file instead of the common VOB file that includes audio and video together. Then took the AC3 file and with another software I converted it from AC3 to 6 WAV channels files (C, FL, FR, LFE, SR, SL). Of which the centered channel includes audio and voices and the other channels only the background audio. The next step and that I have no idea how to do it is to find a program, maybe audacity and tell it that there is one file with both audios, another with only back audio, but that I need it to recognize and remove the back audio so I can have the character voices only. Anyone knows how to do this? Here are sample of the two audios I got:

Back and voices (they are in Spanish)
http://www.mediafire.com/?g5u163eteiru776

Back only (instrumental)
http://www.mediafire.com/?imeudzy1ia65pxx
 
That's the bit that's not possible (as far as I am aware). If the center channel has audio and vocal mixed  you can't separate them. It's a little like trying to get back the flour after the cake has been baked.
The vocal track just doesn't exist as a separate entity - only the vocal/music mix.
At least, every site I have looked at with people answering similar questions say the same thing. Not possible. Anything used would effect both vocals and music (like noise reduction stuff).
I hope I'm wrong. There are a few park recordings (spiels etc) that would be nice to have the background music removed from and some rides that it would be nice to have just the isolated vocals from.
I understand what you want (I think). You want a program that you can load the audio/vocal mix into and then load the music and tell it "remove B from A" but I doubt it's possible without serious cash being spent if at all.
Where's our experts when you need them  :D
If you really want to give it a try
http://www.ehow.com/how_5040761_remove-background-music-audacity.html
Didn't work for me though.
Quite a heated debate about it here:
http://www.audiomastersforum.net/synforum/13/topic-13082.htm
 
Once mixed they cannot be unmixed with any degree of success.    The frequency of the human voice :  The voiced speech of a typical adult male will have a fundamental frequency from 85 to 180 Hz, and that of a typical adult female from 165 to 255 Hz.[

The frequency of orchestral instruments can be found here: http://hyperphysics.phy-astr.gsu.edu/hbase/music/orchins.html

you can see the vast overlapping.  That over lap prevents separation.

 
Well, I believe normally, voice removal is done like this.

If the vocals are not stereo, meaning the same on the left and right channels, you remove the like frequencies from both the left and right channelsqq and what's left would be the music.  You basically AND the left and right and then XOR that from both the left and right..  I think...  My boolean logic is a little rusty.  Even this is pretty destructive to the music, but that's another story.

If the vocals are stereo, then I think you're SOL.
 
Indeed. If vocals are on one track and music on another that`s easy.

If they are mixed you cannot fully remove vocals. You can try to remove the frequencies that carry voices, but since music also uses these too you`ll end up with something bass heavy, a little top and hardly any mid. Certain software may also try to introduce a little reverb too that makes it sound worse.
 
You must log in or register to reply here.
Back
Top