Front Page › Forums › AUDIO & TECHZONE › D.I.Y Acapellas › Making your own acapellas from tracks…
-
AuthorPosts
-
February 1, 2004 at 1:44 pm #529580
TiMGParticipant—–
[b:20ydroug]EDIT BY ADMINS:[size size=150:20ydroug]Please only reply to this thread if you have something to contribute – Dont make useless 1 liner “Thanks” posts![/size:20ydroug]
We all appreciate the hard work TimG has put into this guide, but please keep the thread clean with real posts containing info/questions… Thanks…[/b:20ydroug]
END EDIT
—–ok. Something i’ve been discussing on GYBO with people.
[quote:20ydroug]Tim G’s Pella making tutorial
1. Get ur mp3 (make sure its atleast 192kps , preferably 256 or more) or WAV (much better obviously)
2. Load it into soundforge and do a ‘PAN/EXPAND’ process with the settings set to ‘mid side’ ..
This gives you a WAV with the Left-right material and centre mono material separated..3. Grab the Centre mono material (which will hopefully have significantly more vocal than instruments/music.. save it as a mono WAV and load into Cool Edit
4. In Cool Edit select a section of instrumental with no vox (bit you want to reduce/remove from the vocal) and get a noise reduction profile (at 24000 FFT size + about 300 snapshots in the profile).. **actually since posting this on GYBO i’ve found that smaller FFT sizes with a larger no of snapshots can be more effective**
5. then select the whole WAV.. preview the noise reduction.. listen to what it does..
6. In the eq graph line bit.. increase the noise red below 200 hz and less between 300-10000hz .. giving a bit more clarity to the main vocal ‘region’
7. Then maybe try some midband compression (300-8000hz) with a very fast attack (with readahead set to about 12ms) and 100ms release.. (i’ll have to post a pic of the comp curve to show you how to get good results) to help gate some of the background music left over.
8. Then go to http://www.soundhack.com and download the amazing, yet seemingly unheard of,.. spectral plugins (VST) (theres a free trial)
9. go into cubase (or wot ever prog u use) import the ‘pella’ you’ve made and use the spectral gate to reduce non fundamental harmonics (see soundhack pdf for more info)…
The trick is to do this ‘in the mix’ for best results… Using a fast attack and slow release with the spectral gate helps remove transients like snares and cymbals..
[/quote:20ydroug]This is a method of getting vocals out of tracks I’ve been using for a while. I’ve been able to get some pretty good results from it.. most the tracks I’ve done for MTV have used this method as acapellas weren’t available for the tracks i wanted to mix.
I’m looking for like minded, pella-making-people with applemacs! Theres a plugin for SonicWorx called Pandora RT which claims to isolate and remove vocals (not via a Left/Right phase inversion)..I don’t have a MAc so i cant try it out.
I think if a few people were to join forces on this we could have a kind of ‘pella ‘release group’.. where we would combine our efforts to get popular + non-available pellas made pretty quickly.
AdSense 336x280February 1, 2004 at 1:44 pm #564973
TiMGParticipant—–
[b:20ydroug]EDIT BY ADMINS:[size size=150:20ydroug]Please only reply to this thread if you have something to contribute – Dont make useless 1 liner “Thanks” posts![/size:20ydroug]
We all appreciate the hard work TimG has put into this guide, but please keep the thread clean with real posts containing info/questions… Thanks…[/b:20ydroug]
END EDIT
—–ok. Something i’ve been discussing on GYBO with people.
[quote:20ydroug]Tim G’s Pella making tutorial
1. Get ur mp3 (make sure its atleast 192kps , preferably 256 or more) or WAV (much better obviously)
2. Load it into soundforge and do a ‘PAN/EXPAND’ process with the settings set to ‘mid side’ ..
This gives you a WAV with the Left-right material and centre mono material separated..3. Grab the Centre mono material (which will hopefully have significantly more vocal than instruments/music.. save it as a mono WAV and load into Cool Edit
4. In Cool Edit select a section of instrumental with no vox (bit you want to reduce/remove from the vocal) and get a noise reduction profile (at 24000 FFT size + about 300 snapshots in the profile).. **actually since posting this on GYBO i’ve found that smaller FFT sizes with a larger no of snapshots can be more effective**
5. then select the whole WAV.. preview the noise reduction.. listen to what it does..
6. In the eq graph line bit.. increase the noise red below 200 hz and less between 300-10000hz .. giving a bit more clarity to the main vocal ‘region’
7. Then maybe try some midband compression (300-8000hz) with a very fast attack (with readahead set to about 12ms) and 100ms release.. (i’ll have to post a pic of the comp curve to show you how to get good results) to help gate some of the background music left over.
8. Then go to http://www.soundhack.com and download the amazing, yet seemingly unheard of,.. spectral plugins (VST) (theres a free trial)
9. go into cubase (or wot ever prog u use) import the ‘pella’ you’ve made and use the spectral gate to reduce non fundamental harmonics (see soundhack pdf for more info)…
The trick is to do this ‘in the mix’ for best results… Using a fast attack and slow release with the spectral gate helps remove transients like snares and cymbals..
[/quote:20ydroug]This is a method of getting vocals out of tracks I’ve been using for a while. I’ve been able to get some pretty good results from it.. most the tracks I’ve done for MTV have used this method as acapellas weren’t available for the tracks i wanted to mix.
I’m looking for like minded, pella-making-people with applemacs! Theres a plugin for SonicWorx called Pandora RT which claims to isolate and remove vocals (not via a Left/Right phase inversion)..I don’t have a MAc so i cant try it out.
I think if a few people were to join forces on this we could have a kind of ‘pella ‘release group’.. where we would combine our efforts to get popular + non-available pellas made pretty quickly.
AdSense 336x280February 13, 2004 at 9:54 am #564979
acapellaKeymasterTim, that is an absolutely awesome tutorial, although people may not be replying directly to this thread, I’m hearing many reports of ppl having Huuge success like yourself with this acapella extraction method.
As to your other comment..
TiMG wrote:I think if a few people were to join forces on this we could have a kind of ‘pella ‘release group’.. where we would combine our efforts to get popular + non-available pellas made pretty quickly.This is something that I would like very much to be kept up to speed on.. maybe you’ve had some replies about it already and we don’t know about it, but could you keep me informed personally.. I think it’s an excellent idea.
admin
AdSense 336x280February 24, 2004 at 8:38 am #565002
wcwcParticipanthave u tried creating an acapella by adding the INVERTED instrumental over the top of the original song?
u’d have to make sure the song and its instrumental are exactly the same though, so you need to get them off a cd. cd single’s that come with the song and its instrumental work most of the time.
then convert the original song and it’s instrumental into wav files and open them in soundforge. invert the entire instrumental. then copy the instrumental and “paste>mix”it with the original song.
the two need to be mixed to match up perfectly in timing so use your zoom tool to help u out
if it is done correctly, the inverted instrumental should cancel out the instruments in the original song, leaving just a clear, crisp acapella behind.sorry if u dont know wot the hell im talkin about, i’m kinda shocking at explaining stuff
AdSense 336x280February 24, 2004 at 1:04 pm #565003
acapellaKeymaster[quote quote="wcwc":25z3pmzg]
sorry if u dont know wot the hell im talkin about, i’m kinda shocking at explaining stuff[/quote:25z3pmzg]I think you did a real good job there wcwc. good stuff for everyone to digest
AdSense 336x280February 27, 2004 at 1:41 pm #565006
TiMGParticipantyeah.. the method i’m using is only a last resort when a pella or instrumental (for phase inversion) cannot be found..
Compression curves.
i don’t always do this .. if i do then i do a curve that basically expands aswell as compresses (compands) .so the quiet stuff cuts out and the high end stuff evens out a bit.I’m now doing stuff with st3pan0va’s vst plugin (been discussing over at gybo)
.. its pretty good give it a go. http://www.freewebs.com/st3pan0va
The idea is you get a bit of instrumental from a track (even if it isnt a perfect match) and loop this underneath the vocal sections.. the plugin then performs a spectral subtraction of one from the other…
I’ve got some good results with this.. its still not perfect but I’m going to have a look at the code for it (which he’s sending me i think) and see if i can’t adapt the process better.
It is very good for reducing the drums which is often the hardest thing to do.AdSense 336x280March 6, 2004 at 3:03 pm #565028
akademiksParticipantugh…. too many programs……
i only got soundforge and acid.AdSense 336x280March 31, 2004 at 9:51 pm #565161
MelParticipantI think I’m going to give this a try myself. I’m going to add a bit of my formula in the method. If I get any good results, you guys’ll be the first to know.
As far as that pella release team mentioned in earlier posts, if we come up with a successful way to do that, COUNT ME IN!!! I’d love to help!
AdSense 336x280April 18, 2004 at 8:32 pm #565297
eikcamlloydParticipantTim tried your tutorial it works but still not satisfied with end result, will try again let u know
AdSense 336x280April 18, 2004 at 11:12 pm #565299
eikcamlloydParticipanti tryed separating vocals on a PC with soundforge at a friend studio saving the left audio and then compressing, result still not what were looking for. i have a G4 and cant find a software that is capable of doing this. do;es anybody have any idea;s
AdSense 336x280April 19, 2004 at 1:01 am #565300
resetParticipant
I have used the insturmental inversion techniqe with moderate success, but the insturmental versions often do not exactly match the original song. this process worked fairly well on the cd single for “2-way” by rayvon ft. shaggy. if you want to practice, this may be a good place to start.AdSense 336x280April 21, 2004 at 2:06 pm #565313
wcwcParticipantis this worth a try?
here’s the original + instrumental(inverted) = acapella method
it works with most cd’s ive triedneed proof it sometimes works before trying it?
here is a short snippet of the original song:
[url:3jh1c5ax]http://homepages.ihug.co.nz/~patchett/original.mp3[/url:3jh1c5ax]here is a short snippet of the acapella i turned it into:
[url:3jh1c5ax]http://homepages.ihug.co.nz/~patchett/acapella.mp3[/url:3jh1c5ax]1.) rip the original song and its instrumental to wav files. Make sure they are exactly the same so they will be able to cancell each other out later.
2.) in soundforge (or whatever u use) find a distictive point on each song so that u can match the 2 files up to each other. to make things easier, press ‘m’ where ur at the points so it leaves a marker.
NOTE: the instrumental is the top file in this pic:

3.) double click to the left of a file. everything to the left of ur file should be selected. remove this data (i usually cut and paste it to a new file) and do the same to the other file. ur 2 files compared should look something like this (check by changing ur zoom level)

4.)select what’s left of ur instrumental file go to edit > copy
5.) go to ur original song and click where ur marker is.
6.) go to edit > pastespecial > mix
make sure the ‘inverted’ box is ticked.
play with the ‘volume’ and ‘preview’ so u can cancel as much of the instrumental as possible.

7.) click on ‘okay’
8.) if it is done correctly, it should look something like this:

and that’s it!
sorry about the confusion – hope i havent got u all confused![/url]
AdSense 336x280May 15, 2004 at 5:13 pm #565485
AnonymousInactivethat’s tight
AdSense 336x280May 20, 2004 at 11:15 pm #565533
akbarParticipanttheres a complete turotial on the net that ill give you guys the link too. if you do have two soundfiles o the same loops and invert one of them it makes absolutely no sound. if you have lyrics on one of the loops it will leave the vocals and cancel out the instrumental . so logically it would work with an instrumental and a copy of the same song i think imma try it with a single and see if it works. heres the link :http://4trak.net/dsico/archives/000082.html
hope it helps
AdSense 336x280May 22, 2004 at 8:35 pm #565572
acheadKeymaster[quote quote="TiMG":1jp0b1k8]ok. Something i’ve been discussing on GYBO with people.
[quote:1jp0b1k8]Tim G’s Pella making tutorial
1. Get ur mp3 (make sure its atleast 192kps , preferably 256 or more) or WAV (much better obviously)
2. Load it into soundforge and do a ‘PAN/EXPAND’ process with the settings set to ‘mid side’ ..
This gives you a WAV with the Left-right material and centre mono material separated..3. Grab the Centre mono material (which will hopefully have significantly more vocal than instruments/music.. save it as a mono WAV and load into Cool Edit
4. In Cool Edit select a section of instrumental with no vox (bit you want to reduce/remove from the vocal) and get a noise reduction profile (at 24000 FFT size + about 300 snapshots in the profile).. **actually since posting this on GYBO i’ve found that smaller FFT sizes with a larger no of snapshots can be more effective**
5. then select the whole WAV.. preview the noise reduction.. listen to what it does..
6. In the eq graph line bit.. increase the noise red below 200 hz and less between 300-10000hz .. giving a bit more clarity to the main vocal ‘region’
7. Then maybe try some midband compression (300-8000hz) with a very fast attack (with readahead set to about 12ms) and 100ms release.. (i’ll have to post a pic of the comp curve to show you how to get good results) to help gate some of the background music left over.
8. Then go to http://www.soundhack.com and download the amazing, yet seemingly unheard of,.. spectral plugins (VST) (theres a free trial)
9. go into cubase (or wot ever prog u use) import the ‘pella’ you’ve made and use the spectral gate to reduce non fundamental harmonics (see soundhack pdf for more info)…
The trick is to do this ‘in the mix’ for best results… Using a fast attack and slow release with the spectral gate helps remove transients like snares and cymbals..
[/quote:1jp0b1k8]This is a method of getting vocals out of tracks I’ve been using for a while. I’ve been able to get some pretty good results from it.. most the tracks I’ve done for MTV have used this method as acapellas weren’t available for the tracks i wanted to mix.
I’m looking for like minded, pella-making-people with applemacs! Theres a plugin for SonicWorx called Pandora RT which claims to isolate and remove vocals (not via a Left/Right phase inversion)..I don’t have a MAc so i cant try it out.
I think if a few people were to join forces on this we could have a kind of ‘pella ‘release group’.. where we would combine our efforts to get popular + non-available pellas made pretty quickly.[/quote:1jp0b1k8]
hey homie, big up to yo for dat man, it looks heavy but i cant get the smae resulsts as you can
anyway peace
AdSense 336x280May 24, 2004 at 8:17 pm #565611
siriusParticipantIts been a long while sice I dabbled with Accapella’s, Ive hardly done anythin with them since the days even before “Classic Accapella’s” was online! lol.
There was (like here) a massive array of different ideas/suggestions about how to do it – I never found a single one to work cleanly, but as I say, its been a good few years since I even tried.
Im looking forward to trying the original suggestion and then trying some of the others on top. For those who are trying to use Cooledit on its own for inversion and stuff like that I dont think it works in the same manner as other softwares – I believe you need to use a handfull of softwares on a single try, using thier strenghths for each task within the sonic process.
I love Cooledit (or should I say Audition Adobe nowadays as Ive found out this week
), but I dont believe it has all the EXACT same methods as other software. I hear on the Adobe site there are new tools coming out in the soon to be released/new version which may be great
I dunno.Well, Just need to lay my hands on soundforge next
Cheers for the tips guys, I sure hope I can cope better as I hadnt much luck last time around! Ive about 2,000 oldskool piano/rave vinyls and about 2200 house/prog/trance to test on!!!!!
Cheerio
Sirius.
AdSense 336x280May 26, 2004 at 6:09 pm #565659
TiMGParticipantyes.. thats a good tip mcs. always look for different mixes of the track before u start and get them as high qual as possible.
<—- Eurythmics ‘Whose that Girl’ extraction of chorus and bridge sects that i did for use in..
unwanted MTV bootleg..used cooledit Noisered and Sp3novas knockout plugin..explained in another thread here somewhere.
AdSense 336x280May 30, 2004 at 12:34 pm #565695
TiMGParticipant[quote:3bwxtog8] dont really understand this:
3. Grab the Centre mono material……..[/quote:3bwxtog8]soundforge. pan/expand with midside preset
AdSense 336x280June 8, 2004 at 6:13 am #565756
djballaParticipanttimg – wow. it’s so frustrating trying to make/create acapellas, etc…so it was so beneficial to see that helpful breakdown. <– i’m stating the obvious, but i wanted to send u my thanks neway. i haven’t finished making one yet but i’m getting there. i’ll let ya know if it turns out alright. peace
AdSense 336x280June 8, 2004 at 10:37 am #565760
TiMGParticipantcheers its nice to be appreciated
AdSense 336x280June 15, 2004 at 9:45 pm #565843
Dj AniciParticipantId like to say i understood some of the things you were talking about……but i didnt, as a matter of fact i had no idea what cooledit was until i saw you posted something similar to this thread on the “dsico” site….if i can get my hands on all the software included in this little tutorial, i’ll definately give this a try
AdSense 336x280August 8, 2004 at 6:45 am #566220
aaj2006Participantwell bbatson, it doesnt matter if they are mono or stereo, just as long as they are both the same thing. i dont think it’ll work with a mono instrumental and a stereo song. just as long as both tracks are either mono or stereo, the inversion should work either way.
AdSense 336x280August 21, 2004 at 1:21 pm #566328
TiMGParticipant[quote:1848indt]When we have a few if someone nice enough has the time could he make a quick website that we could just send, post pellas to you to put on it?? [/quote:1848indt]
i’ll try and set up access to a folder on my ftp so anyone can upload stuff if u like..
AdSense 336x280August 21, 2004 at 1:21 pm #566329
TiMGParticipant[quote:3pe3gcov]When we have a few if someone nice enough has the time could he make a quick website that we could just send, post pellas to you to put on it?? [/quote:3pe3gcov]
i’ll try and set up access to a folder on my ftp so anyone can upload stuff if u like..
AdSense 336x280September 21, 2004 at 9:04 am #566650
TiMGParticipantin soundforge its on the process menu
AdSense 336x280October 18, 2004 at 3:02 pm #566882
AnonymousInactiveYeah …. works great with dirty south acapellas like Lil Jon, Lil Scrappy etc.
BIG THX
AdSense 336x280October 23, 2004 at 4:07 pm #566942
TiMGParticipantsp3nova.. knockout… read wot i’ve put on the threads…. use some imagination
AdSense 336x280November 11, 2004 at 2:45 pm #567106
junoParticipanti’ve been playin around with this method and have gotten some pretty decent results. It just takes some experimenting with different techniques. My DIY’s are getting closer to being quality enough to mix. The only downside is that they sound too high on the treble end…………i’m gonna try to remaster em in Soundforge. Anyways, Thanks TIM
AdSense 336x280December 6, 2004 at 9:51 pm #567536
TiMGParticipant
heres wot the noise reduction bit should look like. use as many snapshots as you can in the profile, press ‘get profile’ (make sure you only select a small area that contains good background ‘noise’ like bass or synth) then apply to the whole thing (preview 1st obviously) (notice that the LOG box is checked .thats important)
theres no rules to what will give you the best pella! u have to use your ears and brain. i rarely use soundhack now, maybe if theres some annoying high frequency left over like a hihat or something i will
knockout is the best thing to use. post or pre cooledit. depends on the song
AdSense 336x280December 7, 2004 at 6:16 pm #567555
TiMGParticipantthe eq graph line bit is the bit where u can see a line with nodes on it in the pic i’ve just posted .
forget about the midband comp. its tricky and i hasrdly ever do it
its called dynamic processing in cool edit.. a compressor basically
theres a tab for min and max frequenciesAdSense 336x280 -
AuthorPosts
The forum ‘D.I.Y Acapellas’ is closed to new topics and replies.
