RE: Vocal Isolation Track (help?)

[monsterjazzlicks]

Hi folks,

I am looking to find an isolated VOCAL TRACK of female singer Evelyn Thomas on her 1984 song: HIGH ENERGY.

I believe these are called ACAPPELLAS?

I have found many sites which claim to offer this, however when you click on the LINK of the song, it diverts you to gambling, financial sites and alike instead. I have been looking on and off since Monday.

I don't mind paying for the track as long as the quality is acceptable.

Or maybe a fellow member on this site has MELODYNE software (or similar) and is able and kind enough to perform the task for a reasonable fee?

Many thanks in advance . . .

Paul David Seaman

[RedLeo]

I believe these acappellas mostly come from video games like Guitar Hero and Rockband, where people have hacked into the software and copied the tracks, and the rest come from people getting hold of copies of the original multitracks of the recording session. Unfortunately, whether an acapella track exists is largely determined by the popularity of the artiste, so I think you are going to have to get someone to do one for you.

However, making one electronically by manipulating the audio data, basically by filtering out everything else, leads to results that are less than perfect, so it depends what you want it for. You're not going to get anything like the quality of ones you've heard - these give you the "pure" original because they are copies of the original isolated studio track, they're not created by processing a fully mixed track.

[monsterjazzlicks]

Cheers Leo,

I appreciate your advice here.

Yes, I understand everything you are saying and I dare say that you are totally correct.

When you look at the likes of MELODYNE's demonstrations on You Tube (or wherever), they always seem to make it look so easy as well as the end result sounding nothing less than perfect. I guess this could be viewed as more of a sales pitch rather than a true representation of the finished product and the amount of time and effort that the end user has had to put into producing it.

Ta,

Paul

[RedLeo]

Melodyne's DNA is a remarkable tool, no doubt about that, but it very much depends on the complexity of the music it's working with. It's a bit like trying to extract the raw egg back out of a baked cake, so - not easy. That's why I asked what you wanted it for. You may be able to get results good enough to build a new arrangement round the extracted vocal, but it would depend heavily on the fact that other instruments will go a long way to covering up the shortcomings of the extracted track - assuming you can get something workable in the first place. I have only seen it done reasonably well where the original arrangement was just one acoustic guitar and the lead vocal, although I certainly haven't seen every video and tutorial out there, and I honestly don't know where the limits of DNA's capabilities really lie.

But it's well worth a shot, you've certainly got nothing to lose.

[monsterjazzlicks]

Cheers Leo,

Well I tried to post a thread on the MELODYNE Forum, but it no longer exists! So I emailed their Support Team (who replied immediately and informatively) telling me that hardly anyone ever used the Forum over the past 18 x months, and so it was closed down!

I think it will be easier if I give you a quick explanation of my scenario (so as to help cut down on your time and brain power!). I did not go into too much detail in my OP because I was not sure if anyone would respond or not:

I have downloaded and imported an INSTRUMENTAL version of "High Energy" into Cubase. If you will remember, in the 80's, singles often had a version of the 'A Side' track on the 'B Side' but without the vocals! This seems to be so for "High Energy". Therefore, there should be no clashing or harmonies or rhythms. In other words, I would assume that each version would sit and play alongside each other (when both audio files are imported and lined up within Cubase)?

Many thanks,

Paul

[RedLeo]

What are you actually trying to do? I'm afraid I'm none the wiser here.

The reason I'm asking is because there might be a very simple way to do this. If the instrumental mix is EXACTLY the same mix, just without the lead vocal, then you can do this:

Line up the two versions in Cubase, and then just invert the phase of the instrumental mix. Make sure the volumes are equal, and all the backing tracks should more or less cancel each other out, leaving you with just the lead vocal. It won't be perfect, there will be some artifacts, but it may well be good enough.

[monsterjazzlicks]

Thanks Leo,

Ok, so I want to have a bash at making my first ever REMIX of a song. I have never done anything like this before and so it is a totally new venture.

I plan on making SLICES and LOOPS from the INSTRUMENTAL version (in Cubase). Then I hope to RE-ARRANGE these chunks to form the basis of a new structure. After that, my aim is to layer a couple of hardware synths over the top (eg. Korg Poly 61 and Yamaha DX7); these will either duplicate material on the track and/or add new textures/motifs. Finally (if I get that far!), this is where I was hoping to obtain a pure melody vocal track so that I can chop it up and manipulate it as I see fit for the purposes of the project.

It probably won't be anything like this (though I wish it could be!), but here is an example of the type of sound I have going on in my head (I just don't know how to get it out if ya' know what I mean!):



'Invert Phase' = is this an AUDIO PROCESSING function within Cubase please? From what you are kindly suggesting, I am not sure if this will work because the two versions will be radically different (in both length and order of events).

Thanks a lot for showing an interest into my disturbed mind!

Best,

Paul

[RedLeo]

'Invert Phase' = is this an AUDIO PROCESSING function within Cubase please? From what you are kindly suggesting, I am not sure if this will work because the two versions will be radically different (in both length and order of events).

Yes, it's a simple audio processing function.

I'm confused because I thought you said the instrumental version was the same as the original version with the lead vocal, now you're saying it's different? Well, in any case, if it's not the same, the idea won't work.

[monsterjazzlicks]

Hi Leo,

Just to ask/clarify, what I mean here please is:

If I cut/slice etc the INSTRUMENTAL version of the song (hence creating a unique version), and then IMPORT the version with Evelyn Thomas's VOCALS into Cubase, obviously they will not match in length, form, or otherwise.

Therefore, if I want to use the INVERT PHASE facility, I will have to IMPORT both the INSTRUMENTAL and VOCAL (original) versions of the song as the very FIRST step of the whole process. Then, once the PHASE INVERSE have been applied (to the VOCAL track), I can cut/slice freely as desired.

I think that would be the method you are suggesting?

Cheers,

Paul

[RedLeo]

Yes. You would have to use the original instrumental mix to do the phase-invert thing. Once you've got a workable isolated vocal track, then have at it with the creativity. Apart from anything else, it would be pointless to invest time and effort into making your own arrangement before you knew for sure that you were going to be able to have a workable isolated vocal track in the first place.

[Remember, in Cubase you will be working non-destructively. What that means is that you will produce a new, third track containing only the vocal without altering the two original tracks in any way].

The phase-invert trick only works if the two things are identical to each other in every respect, except of course for the stuff you want to keep.

If you want to know how it works, here's a crude explanation:

It works on the simple principle that +1 plus -1 equals zero.

If you take your original track to be "+1", then the inverted track becomes "-1". Add the two together at equal volume, and you've got a result of zero - the two "opposite" tracks have perfectly cancelled each other out, leaving you - theoretically - with complete silence.

Here's the clever bit - the vocal is only on ONE of the two tracks. So the inverted instrumental track does NOT invert the vocal (because there isn't any vocal) so NO inverted vocal is added to the original track. Thus the original vocal part is left untouched - hey presto.

Edit: here's a video tutorial which explains how to do it: