Acapellas4U.co.uk

usefull info, thx

Hi I'm a newbie here, but I have been visiting this forum for a while.

As mentioned before in this topic, Celemony's autotuning software Melodyne will release a polyphonic autotuner. Today the beta version has been officially released. See how it works here:
http://www.celemony.com/cms/index.php?id=editor_betavideo

Some beta testers has allready tried to split up mixes with some succesful results.
Check out their experiences at the Celemony forum:
http://www.celemony.com/userforum/viewtopic.php?f=21&t=4963

As I haven't a Melodyne license myself, I have not been able to check how good results you can get.

thank u mate, great stuff

Synthetik FM wrote:I discovered another method: you need two mixes of the same song, completely different except for that the vocals need to be exactly identical. Line up both mixes in acid, make sure the vocal lines up (you should hear a flange), hard pan left and right, render a wav.

Do a phase cancellation on the mix to eliminate the vocals. I use Nero Wav Editor's karaoke Filter, which will basically create instrumetal versions of each mix in each channel. Choose the channel you want to use, then line up the instrumental mix versus the original mix in ACID again. Render a wav, do phase cancellation again, and you have isovox!

did anyone get this method to work? i tried it with a song and its remix but i got no dice. i'm hoping Synthetik FM can come back and supply us with a bit more detail. which channel is the one after applying the kareoke filter that i want to use, which ever one is more devoid of vox right? what other progs, windows or mac, have a phase cancellation option, is that the same thing as flipping/inverting in sound forge/audacity (respectively)? do i bounce the channel from the kareoke filter i want to use as a stereo image, i.e. copy the one channel to both channels, or do i just leave it in L or R when i line it up in Acid?
it's the last phase cancellation that's really getting me tho. reapplying the kareoke filter seems to have the opposite effect no matter how i conduct the overall process, it almost seems to make the song whole again, at least in one channel...
plz, lil' help, if you're still there duder, or anyone who got this method to work? much obliged..

i don't beliwe on it

Did know to method 1 but not 2 and 3 so thanks

saaaaaaaaaaaaaaaaaafe man

about the only ways to make DIY-acapellas but they'll never be the same as the original though.

Good overview Tim!

Method 3 works much better now that Audition has upgraded it's processing abilities. There is also a Center Channel Extractor in the Stereo Imagery menu that will help isolate portions of the audio. Play with the FFT Filter and Noise Reduction to eliminate any remaining/unwanted sound

really cool tip that mate . never thought of doing it like that

<< Newb here...read through the topic and did some searches here. Is it safe to say that if you do not have an instrumental of a song and/or the song is not completely looped the same that your chances of getting a acapella of it are slim to none?

Example of this would be I am looking at getting the vocals of Breaking Benjamin - Without You. There is no "instrumental" portion of the song that is the same at the chorus/hook so I would not be able to use the "knockout" method on this part.

Is there anyway to go about taking a entire track and cutting the vocals out of it and then inturn using that "instrumental" version you have just created to invert it against the original track?

UneikSoundz wrote:Example of this would be I am looking at getting the vocals of Breaking Benjamin - Without You. There is no "instrumental" portion of the song that is the same at the chorus/hook so I would not be able to use the "knockout" method on this part.

I have been quite sucessful in KnockOut verse 1 with verse 2, or chorus 1 with chorus 2, if the singing is the only difference between them.

UneikSoundz wrote:Is there anyway to go about taking a entire track and cutting the vocals out of it and then inturn using that "instrumental" version you have just created to invert it against the original track?

There are some vocal remover/kareoke filters out there that remove certain frequences of the stereo centre. Sometimes you got a quite decent instrumental. Unfortunately, trying to KnockOut the original with the created instrumental has never been successful to me. You got to much muddy noice with the vocals in the end.

For me, I tried the pahse cancellation method, didn't work at all. I'm using
Audacity(updated beta ver.). I recently stared using Voice Trap and Extraboy.
The Extraboy demo looked promising but I'm sticking with Voice Trap.
The thing with VT is some tracks work better than others, I'm working with
different NIN tracks from Pretty Hate to The Fragile. In the beggining, I got
frustrated and discouraged but I'm still hacking away at it.

As far as the Extraboy VST, I didn't like how it extracted the vocals, it chopped
the hell out of it, whatever. I might not get an exact acapella but the results are
too interesting to ignore, that's my take on it.

If anyone is interested, I'll post the addy for both cause right now it's a total
blur.

Ah so there are methods to take out acapellas and instrumentals! Awesome tips! ;)

thx

dwnescp wrote:

Synthetik FM wrote:I discovered another method: you need two mixes of the same song, completely different except for that the vocals need to be exactly identical. Line up both mixes in acid, make sure the vocal lines up (you should hear a flange), hard pan left and right, render a wav.

Do a phase cancellation on the mix to eliminate the vocals. I use Nero Wav Editor's karaoke Filter, which will basically create instrumetal versions of each mix in each channel. Choose the channel you want to use, then line up the instrumental mix versus the original mix in ACID again. Render a wav, do phase cancellation again, and you have isovox!

did anyone get this method to work? i tried it with a song and its remix but i got no dice. i'm hoping Synthetik FM can come back and supply us with a bit more detail. which channel is the one after applying the kareoke filter that i want to use, which ever one is more devoid of vox right? what other progs, windows or mac, have a phase cancellation option, is that the same thing as flipping/inverting in sound forge/audacity (respectively)? do i bounce the channel from the kareoke filter i want to use as a stereo image, i.e. copy the one channel to both channels, or do i just leave it in L or R when i line it up in Acid?
it's the last phase cancellation that's really getting me tho. reapplying the kareoke filter seems to have the opposite effect no matter how i conduct the overall process, it almost seems to make the song whole again, at least in one channel...
plz, lil' help, if you're still there duder, or anyone who got this method to work? much obliged..

Hey ... sorry for being gone for awhile. The idea of this is you are using both mixes to get an instrumental version of one of the mixes, which you then overlay back with the original track to re-isolate the vox.

An example is my acapella for ABC's "The Look of Love", using both the original 1982 mix and the 1990 mix. It's available in the acapella library on this site. Here's what I did:

Take both mixes, line them up in acid and beatmap them, make sure the vocals flange when they overlap. You might have to alter the start offset on one of the wavs.

Hard pan each mix left and right, so you get a different mix coming out each side.

Render a wav and use phase cancellation in Nero Wav editor using the karaoke filter. This is to knock out the vocals from both sides of the mix. You should end up with a mix featuring near instrumental versions of both songs, one out each side.

Choose one side of the mix, and in audacity, copy this side of the mix to the other channel. In my case I chose to use the original 1982 mix, which is now in near instrumental form out both speakers after you do this step..

Now take this wav back into acid, overlaid with the original 1982 vocal mix. hard pan left and right on both wavs, render a wav again.

Take the wav you just rendered, which should be instrumental out one side, vocal out the other, and reapply the karaoke filter in Nero Wav editor.

This should make the acapella.

i would like to apologize for my old comments. i was a dumb ass young kid who thought everything he said was right. haha. how the funk can i be so dumb. that was years ago. i did do the voice imitation thing, it worked. but i didnt realize that every song made has to have a master acapella. thats how u double up on tracks. i guess when ur young u dont realize how songs are made. u think that theres a program to make ur voice sound like theres 2. yea i did figure out what i was hearing in the backround of emimiems song with 2pac. in ghetto gospel i think it was, the asshole slowed down his voice like 60 or 70% and made it sound all chipmunk'y. i was right about one thing that master acapella sounds like a different cut. i always wondered why they never used the original cut, and the other 2 verses. matter fact em just funked that whole album up. anyway sorry for sounding like a dumb ass haha. i was young and dumb.

I've managed a decent one to C-Ride's "Money Round Here (feat. T-Pain)" but you can still hear too much of the music in the background. I've been trying EQ's and filters to help but alas, nothing. I'm going to try some of these other options everyone keeps going on about and see if I can better results. I must say I underestimated the difficulty of producing one of these.

uptheirons106 wrote:I've managed a decent one to C-Ride's "Money Round Here (feat. T-Pain)" but you can still hear too much of the music in the background. I've been trying EQ's and filters to help but alas, nothing. I'm going to try some of these other options everyone keeps going on about and see if I can better results. I must say I underestimated the difficulty of producing one of these.

The more the tracks are compressed during the mastering process, the more the instrumental-part-in-the-original differs from the clean instrumental. And that leaves a lot of blurry background noice in the diy acapella.

I mite try this out. thanks for the tips!

[quote="TiMG"]heres an overview of the different methods used for extracting vocals...

1. THE PHASE CANCELLATION METHOD
Get an instrumental of a song, invert the phase (sometimes refered to as flipping), and mix it with the original.. If done right everything besides the vocal is cancelled out. You can do this in any sequencer like cubase or acid.. The instrumental must be excatly the same time/pitch however. and mp3s might not work if they are badly encoded. Zoom right in to see the 2 waveforms next to each other (look for kick drum hits) and line them up.

TimG[/quote]


Doesn't this method isolate the instrumental?

cbuffalino wrote:

TiMG wrote:heres an overview of the different methods used for extracting vocals...

1. THE PHASE CANCELLATION METHOD
Get an instrumental of a song, invert the phase (sometimes refered to as flipping), and mix it with the original.. If done right everything besides the vocal is cancelled out. You can do this in any sequencer like cubase or acid.. The instrumental must be excatly the same time/pitch however. and mp3s might not work if they are badly encoded. Zoom right in to see the 2 waveforms next to each other (look for kick drum hits) and line them up.

TimG

Doesn't this method isolate the instrumental?

MAYBE

can u extract vocals from a song on reason 4??????????