Whats the best way to change a song into an instrumentalsong

[Tranman409]

I been trying to find an instrumental version of Coolio's Gangstas Paradise and having no luck since its a 10 year old. I found the originial song and it has the words in it. I really need to do trhis today because this karaoke thing is due tommorow for an economics class hehe. I remember I had an old MP3 player that you could change the sound to "karaoke" and it tried to reduce all the singing in it and it worked perfectly. Please help guys

[housepig]

you can try Vocal Remover from AnalogX, but like most of the software methods, it's going to depend on a) a really dry vocal that's b) mixed mostly in the center, without c) a lot of stereo effects on it.

most results are dissapointing. YMMV. don't wait until the last minute if you want good results.

[Doramius]

There's a nice instrumental opening that can be looped very evenly and the vocal "aww" can be found at the end with no instruments. That can also be cut and dubbed into the loop. THat's about the best I can offer. Just need a digital audio editing software to do it. Preferably once with an oscilloscope or oscillating audio image. You can run you loop of the instrumental 3 or 4 times then add in the "awww" sound exactly to match the rhythm. It's not really like karaoke, but it'll be very close. There are many songs you can use that method with and not worry about cutting out the human voice. Many music creations today have odd effects that blend in or chorus' that are wanted, but get cut out too due to human voice detection. It's not a fast process in any case. My suggestion is take the F and work on the project sooner next time.

[bazooka]

There is no best way.

[BJ_M]

There's a nice instrumental opening that can be looped very evenly and the vocal "aww" can be found at the end with no instruments. That can also be cut and dubbed into the loop. THat's about the best I can offer. Just need a digital audio editing software to do it. Preferably once with an oscilloscope or oscillating audio image. You can run you loop of the instrumental 3 or 4 times then add in the "awww" sound exactly to match the rhythm. It's not really like karaoke, but it'll be very close. There are many songs you can use that method with and not worry about cutting out the human voice. Many music creations today have odd effects that blend in or chorus' that are wanted, but get cut out too due to human voice detection. It's not a fast process in any case. My suggestion is take the F and work on the project sooner next time.

that would be called SONY ACID, Image-Line Software Fruity loops, Propellerhead ReCycle, Cakewalk Sonar, Plasma, BitHeadz Phrazer or Ableton Live -to do that ..

[Tranman409]

Thanks guys for the help. gonna try them out

[ViRaL1]

There is no best way.

Not true. The best way is to remove the vocals.

[bazooka]

There is no best way.

Not true. The best way is to remove the vocals.

Or to not listen to the song at all.

[ViRaL1]

There is no best way.

Not true. The best way is to remove the vocals.

Or to not listen to the song at all.

Then it's not instrumental either.

[bazooka]

There is no best way.

Not true. The best way is to remove the vocals.

Or to not listen to the song at all.

Then it's not instrumental either.

True, but with so many programs on hand to do this, my point was that there is bo best method to do it.

[lordsmurf]

Unless you have vocals on separate tracks, I'm under the impression it's pretty much impossible. The music and audio crossover each other. Sort of like removing interlace ... a mess.

Oh, and moved to AUDIO forum.

[BJ_M]

Unless you have vocals on separate tracks, I'm under the impression it's pretty much impossible. The music and audio crossover each other. Sort of like removing interlace ... a mess.

Oh, and moved to AUDIO forum.

it can be done -- in fact there is both software and some fairly expensive hardware to do it -- superscope also makes a commercial preamp/mixer/amp that does an excellent job for not a lot of money..


how WELL it turns out is dependent on a number of things ...


lot of times -- its just easier to play it yourself and record it ..

[Captain315]

It's like cream in coffee.... Once in, it's in. You cannot effectively remove vocals unless they are on a track by themselves.

[guns1inger]

I'd argue that the vocals can often be totally removed (unless the track is mp3, in which case it is very difficult), but you often also remove much of the rest of the track. Software vocal removal applications work by flipping the wave form of one channel, using this to cancel out anything the same in the other channel, then flipping it back. Any sound with the same wave forms in both channels will be removed - vocals, bass, drums. It is totally mix dependent.

[Doramius]

What the traditional systems and software do is take the standard human vocal range and time delay briefly the wave and repeat it causing noise cancellation of the audio. You can do it with any audio source, digital or analog. The problem is some parts of instrumentation will be phased out as well. anything thet falls within the human vocal range. It often catches saxaphone & oboe, even some percussion. Many old creations of music ran instruments and vocals on different tracks. A cassette tape (more recent before CDs took over) may have 4 or 8 tracks in each direction. Each track may have it's own instrument or vocals. If you block out the track then all you have is instrumental. That phased out because it's traditionally simple stereo. Even then sometimes they ran some vocals together with instruments and you have to cancel out the human voice. You effectively take out the human voice, but it takes time and you have to have the right equipment/software. My major in college was broadcasting communications. We worked heavily with this stuff. I hated it, and always took the simple route and cut out as much of the instrumental parts as I could find and respliced them together in a mix close enough to call it instrumental. Most radio stations run loops like these as background for the DJs. They also talk over parts of the song (especially instrumental) so that if you copy the music, the radio station has the right to sue you. It's a deterant for copyright. Some radio stations boast that they don't talk over their music, but they often chop the song early or reduce and instrumental portion. I heard "Come on baby light my fire" a whole 3 minutes shorter than it's actual length and nobody else listening noticed the difference. (BTW - Excellent song DJs play so they can go to the bathroom. Same with American Pie) The song I chose to turn instrumental was NIN "Closer". My teacher made a remark about how the vocal version for comparison was a little "explicit". I got an 'A' though. I wish I still had that instrumental though. I used the tape over again for another project and recorded over it. Oh well.

[alsyed]

i have the instrumental track that came on the cassingle. It had 3 remixes and that instrumental. I'll have a look for it.

[Jester700]

Just a couple points:

What the traditional systems and software do is take the standard human vocal range and time delay briefly the wave and repeat it causing noise cancellation of the audio. You can do it with any audio source, digital or analog. The problem is some parts of instrumentation will be phased out as well. anything thet falls within the human vocal range. It often catches saxaphone & oboe, even some percussion.

This is a little misleading. It's not exactly a time delay, it's a phase reversal. Also, the vocal range has harmonics over the entire range, to over 10k. So the only bandwidth limiting is done in the bass; frequencies below a certain point are passed through unaffected. This can obviously be set higher for sopranos and children (where a 160Hz rolloff often works) than for basses (where you need to go down to 80 or so).

Many old creations of music ran instruments and vocals on different tracks. A cassette tape (more recent before CDs took over) may have 4 or 8 tracks in each direction. Each track may have it's own instrument or vocals. If you block out the track then all you have is instrumental. That phased out because it's traditionally simple stereo. Even then sometimes they ran some vocals together with instruments and you have to cancel out the human voice. You effectively take out the human voice, but it takes time and you have to have the right equipment/software.

I think you're mixing up formats here. Consumer cassettes have 4 tracks - 2 per side. 8 track tapes had - you guessed it - 8 tracks. Many old open reel machines were also 4 tracks, either for 2 sided play or in some cases for quadraphonic sound. But none of this is the same as studio 4 and 8 track machines, where the tracks are individually mixable as you say. Consumer formats were in stereo pairs of tracks, and you couldn't adjust individual elements except in one case... That case was the old 60's recordings (like the early USA Beatles releases) with vocals on one side & instruments on the other. You can easily adjust the mix here with your balance control, of course. But it was all an error - according to George Martin, those were never intended to be released that way but instead mixed to mono. So, why did other 60's groups emulate this unnatural sound? Probably because of the Beatles' rabid success with it.
[/quote]

[theDruid]

GoldWave has a filter to remove vocals (all it's doing is cancelling out the stereo channels against each other as other people have described.)



Sometimes it works okay, sometimes it doesn't really work at all (depending largely on the way the track has been mixed I think, as others have said).

You end up with a mono track (albeit with the same track copied to L & R so it's still a stereo file).

As someone else said too - it doesn't work well with mp3's - it creates a lot of phasing/flanging type noise. You'd have to do it on a raw wav to give it a decent chance of working.

Might be worth a try...

cheers,
theDruid.

[lordsmurf]

<nevermind, found my answer>

[wonderuss]

I also have the original it's on the cd single

[Doramius]

Just a couple points:

What the traditional systems and software do is take the standard human vocal range and time delay briefly the wave and repeat it causing noise cancellation of the audio. You can do it with any audio source, digital or analog. The problem is some parts of instrumentation will be phased out as well. anything thet falls within the human vocal range. It often catches saxaphone & oboe, even some percussion.

This is a little misleading. It's not exactly a time delay, it's a phase reversal. Also, the vocal range has harmonics over the entire range, to over 10k. So the only bandwidth limiting is done in the bass; frequencies below a certain point are passed through unaffected. This can obviously be set higher for sopranos and children (where a 160Hz rolloff often works) than for basses (where you need to go down to 80 or so).

Many old creations of music ran instruments and vocals on different tracks. A cassette tape (more recent before CDs took over) may have 4 or 8 tracks in each direction. Each track may have it's own instrument or vocals. If you block out the track then all you have is instrumental. That phased out because it's traditionally simple stereo. Even then sometimes they ran some vocals together with instruments and you have to cancel out the human voice. You effectively take out the human voice, but it takes time and you have to have the right equipment/software.

I think you're mixing up formats here. Consumer cassettes have 4 tracks - 2 per side. 8 track tapes had - you guessed it - 8 tracks. Many old open reel machines were also 4 tracks, either for 2 sided play or in some cases for quadraphonic sound. But none of this is the same as studio 4 and 8 track machines, where the tracks are individually mixable as you say. Consumer formats were in stereo pairs of tracks, and you couldn't adjust individual elements except in one case... That case was the old 60's recordings (like the early USA Beatles releases) with vocals on one side & instruments on the other. You can easily adjust the mix here with your balance control, of course. But it was all an error - according to George Martin, those were never intended to be released that way but instead mixed to mono. So, why did other 60's groups emulate this unnatural sound? Probably because of the Beatles' rabid success with it.

[/quote]

You're correct about phase reversal, I just felt like explaining it a little easier and not going into the long tanget you used and take up even more space to an alreaday long response. Time delay would be simple enough to get the point across and be done with it.

Also, newer cassettes offered up to 16 tracks total meaning a possible 8 in each direction. However, I don't think any made the commercial market for music sales as CDs had already made it big. This is why I said 4 or 8 tracks in either direction. I also stated that they phased out some of the ways it was done. They also have R2R up to 16 tracks, but nut highly used. Digital is widely more accepted. I've visited some radio stations recently and most only had 1 or 2 4-track R2R players that are just gathering dust. Only one station still had the tape-carts and that machine was also gathering dust. all of them had CD-carts & mini-disc machines integrated with a computer system. They all still had a pair of turntables and cassete decks, but again, they're all gathering dust. The only thing remotely close that was still used on off occasion was the DAT machines, and from what I understand, it's rare.

[BJ_M]

DA88 (or like them) machines are very common still - kinda of a standard in fact

R2R can be more than 16 tracks and analog R2R is still VERY popular in many studios ..

8 track cassete ? you mean digital (see above) or that older (1994) portastudio tascam 688? no longer made ..

or a 8 track minidisk format by yamaha ?

[Jester700]

DA88 (or like them) machines are very common still - kinda of a standard in fact

R2R can be more than 16 tracks and analog R2R is still VERY popular in many studios ..

8 track cassete ? you mean digital (see above) or that older (1994) portastudio tascam 688? no longer made ..

or a 8 track minidisk format by yamaha ?

I thought he was saying there were multitrack consumer formats out there. Maybe I just read it wrong; if so, my bad. Yeah, there were lots of formats to pick from. Akai even made a 12 track on 1/2" Beta videotape; it was an analog recorder...

The point was, though de-voxing is easy in the studio, it's almost impossible for stereo media, and when it works, it still doesn't work very WELL.

[Doramius]

I believe you can call the record label and have the instrumentals sent to you, for a cost, on many songs. I know this is another method radio stations have used for background music, but don't do very often because of the cost. If they can mix it on their own, it's cheaper for them.

Now I'm aching to listen to NIN again. Haven't heard them in a looong time.

[Jester700]

I believe you can call the record label and have the instrumentals sent to you, for a cost, on many songs. I know this is another method radio stations have used for background music, but don't do very often because of the cost. If they can mix it on their own, it's cheaper for them.

Wow. Never heard of that. Only the label has access to he original master, and unless they do a mix specifically for karaoke use, it ain't gonna happen.

The VAST majority of karaoke tracks are done after the fact - companies pay studio players to re-create the song from the ground up. How good this is depends on the quality of players and the time & money spent getting it to sound just right. Some are a little dodgy, and some are frighteningly good. But these are only available for popular tunes, because those are the ones most likely to sell enough to recoup the costs in recording them.

Lots of CD stuff available here:
http://www.karaoke.com/cgi-bin/WebObjects/Karaoke.woa/wa/Karaoke/default
And downloads for a fee available here:

www.singingstation.com

[BJ_M]

studios will not send you the instrumental track unless you pay huge huge bucks -- even sampled stuff is often re-created or sampled partially ..

maybe for radio use you can get a jingle out of them -- but certainly not for any commercial use -- i used to be sent all those mixed for radio cd's (i must have 100's and 100's of them) .. usually compressed to shit even more than the often lame retail versions... and edited for content and lenth of course....

[Doramius]

The cost to get instrumentals from a label is certainly not worth it. Like I said, radio stations don't do it very often. Also, you can't get every song you want as an instrumental. Sometimes they have it and other songs they don't. The karaoke companies usually make their own mix that a load of bull 3/4 of the time. once in a while a good one comes about, but very rare.

[JimJohnD]

Have you checked out a MIDI version. With the newer sound cards you can replace the GM soundbank with a more convincing set. Some of the MIDI versions do have that "mu-zak" sound, but some are really well done. It may not have the exact same sound as the origional but it's easier than trying to unmix paint (or coffee)

[Doramius]

MIDI is just another creation by someone else. That was discussed earlier. I know many Karaoke discs that use that technology for a good creation, then do a conversion to wav for disc printing.

[Manokat]

I been trying to find an instrumental version of Coolio's Gangstas Paradise and having no luck since its a 10 year old. I found the originial song and it has the words in it. I really need to do trhis today because this karaoke thing is due tommorow for an economics class hehe. I remember I had an old MP3 player that you could change the sound to "karaoke" and it tried to reduce all the singing in it and it worked perfectly. Please help guys

I assume you only have two channel stereo in which you have the human frequency (vocal) mixed with the accompaniment on both channels. So essentially you are trying to make a "minus-one" music plus melody out of a complete arrangement off a CD/tape track.

Normally human voice generates around 500hz. Which is located in the center of the range. Unfortunately the melody and harmony parts of some accoustics fall into that range as well. This is called the mid range. So since the channel is limited to only stereo (two), the voice is then downmixed with the melody and harmony of the instruments.

What others have said is actually worth trying. You could try to find a midi file for that song and start from there.

Lastly get any PSR Yamaha keaboard and play your own....