I don't know if this will work, but I want to try taking audio and if possible removing the voices and sound effects so I'm left with only the musical score. I have Audacity and it's supposed to be able to (in certain cases) separate audio clips. However, I ripped the audio from my DVD and it is an AC3 file. Every time I try to load it into Audacity, the program crashes. I'm worried that if I try to convert the file, it will flatten it in such a way that I won't have a chance of removing the unwanted sounds anymore.
I know there are programs that do this sort of thing, but I believe they are quite expensive. If not Audacity, will Final Cut Pro do anything like that? (I also have Audio Hijack Pro, but I'll bet that won't do it) If not, what other options do I have if any?
+ Reply to Thread
Results 1 to 13 of 13
-
-
Audacity won't process AC3 audio, so you have to convert it first. I would suggest to uncompressed PCM, as any sort of lossy compression (MP3, for example) makes further processing difficult.
Then you need to accept that you cannot do what you are attempting to do.
There is software that can be used to remove, in some instances, vocals and effects, however the situations under which it works are very limited. The process is one of audio wave form elimination. If you take a stereo track, invert the wave form, and overlay it on the other channel, anything that is the same in both channels will cancel itself out. IF (and it's a big IF) the vocals and effects are recording equally down the centre of the two channels, and nothing else is, then the vocals and effects will be removed.
This can be effective for some music tracks. It rarely, if ever works on movie tracks because they are never recorded in this manner. And it cannot be used for single channel clean ups, which is what you are looking at on a multi-channel audio track, unless you can sample just the effect to be removed. Audacity's noise removal filter works like this. You isolate a section with just the noise, sample it, and audacity then uses this to apply an inverse version of itself and remove the noise. It works fine for most hiss and hum type noises. Not so effective on single effects, vocals etc where the noise is not regular.
There are other tools that can be brought into play, such as eq's etc, however the more you chip away at the track to hide the effects, the less there is of what you were originally after.Read my blog here.
-
Sorry to bump an old thread, but I haven't had much time to work with this until now.
First off, what are some good programs that can convert to PCM on a Mac? I have various programs, from ffmpegX to Audio Hijack Pro to Audacity 1.2.6 to MPEG Streamclip, but I'm not sure if anything I have can do PCM.
Is it really that rare for it to work with movie audio? I know it isn't easy, and I could swear that the programs cost a bundle, but I have a couple movie scores that I downloaded that other people did where they removed the sounds and vocals (one of the scores still had some sound effects that couldn't be removed, but the other seemed pretty perfect). Maybe the people who did the second of the two had better programs/equipment? -
It depends on who did them, what they were working with, and how much trouble they were willing to go to.
Some DVDs have audio tracks that are just the music or just the foley and vocals, so no separation is needed at all. Somewhere out on the net is a version of the faux James Bond film Never Say Never Again that has been recut and an entirely new soundtrack added that uses John Barry scores. In this case, the fan (lunatic ?) who put it all together actually had to rebuild the foley himself for some scenes because he couldn't separate the music and vocals/effects.Read my blog here.
-
Originally Posted by guns1inger
-
Here is the site for the 'Never Say McClory Again' edit : http://www.ohmss-007.com/neversaymccloryagain/index.html which has samples and link to download the finished work. It also has a tutorial on creating your own alternate soundtracks, which might help your cause.
Read my blog here.
-
Originally Posted by Jeikobu
It's not possible to do perfectly, but look at http://audacity.sourceforge.net/help/faq?s=editing&i=remove-vocals and http://audacityteam.org/wiki/index.php?title=Vocal_Removal for some methods you can try.
Originally Posted by Jeikobu
Use Aften to convert the edited wave to a new AC3. -
In the past, I've used mac3dec http://www.versiontracker.com/dyn/moreinfo/macos/14479
It'll convert to AIFF just fine. I'm sure there are other tools that do this, but that's what I've used. It's free, and it seems to work without complaint.
One problem you might encounter is that it has not been updated in a long time, and may not work with the latest versions of OS X. Other options include ffmpegx (shareware, but a bargain; it's the "swiss army knife" of mac multimedia). Also, VLC's transcoding capabilities may also include conversion of ac3 into wav (I've never tried it, but since VLC is free, you get at least your money's worth!). -
Originally Posted by tomlee59
Originally Posted by tomlee59
Originally Posted by tomlee59
Now that it's a WAV, do I just proceed to edit it in Audacity? -
Congratulations on getting it into .wav format. From here, you can do just about anything that you want. That does include letting Audacity have a whack at it, among other tools. Using Audacity to suppress (though not eliminate) vocals is straightforward. Be aware that you will be removing everything that is common to both channels, and that the result will be a monaural track.
The basic procedure is this: Invert one of the channels (either left or right; doesn't matter which you choose). Then mix (add) the two together to produce a single channel. If you want to do something vastly more sophisticated than this, you'll have to do a great deal more work (with no guarantee of success). For example, if the sound effects you referred to in your OP are not truly common to both channels, then they will not disappear. Also, if you want the result to be in stereo, you have a huge amount of extra work to do. An expedient is to create "pseudo-stereo" by taking your mono track, duplicating it, and delaying one version by about 50ms or so. The small delay will create the impression that the sound is in stereo. Not true stereo, of course, but it's the biggest bang for the buck. You may also wish to provide some bass boost, since low frequencies tend to be common to both channels, and thus tend to disappear during this processing. A bit of bass boost can improve the results markedly.
Similar Threads
-
Separating out a voice in the crowd
By and in forum EditingReplies: 1Last Post: 8th Dec 2011, 08:20 -
Separating Audio Layers From .AVI File
By PhoenixOrion602 in forum AudioReplies: 2Last Post: 5th Jan 2011, 23:58 -
separating audio and re-authoring
By joecass in forum Newbie / General discussionsReplies: 1Last Post: 23rd Dec 2010, 10:37 -
Please Help - Corel VideoStudio Help Separating Videos on Timeline
By djadams88 in forum Newbie / General discussionsReplies: 2Last Post: 15th Dec 2009, 11:49 -
separating audio files
By lordwade in forum AudioReplies: 1Last Post: 11th May 2007, 03:32