questions on how to sample...

[djvinivicious]

Can anyone help me out here, i would like to get just the vocals of a song but not its beats so i can place that on my set, can someone help me here...

I guess i need to know how to separate the bass from the drums from the lyrics in a song and the best thing i found was the DJ Master channel where you can turn up and down the Lo Mids and Highs but it is not quite what i'm trying to do.

any feedback will be appreciated!

[weeddigger]

Have you ever tried un-baking a cake?

You know, separating the eggs from the flour and the cake mix after they been mixed into batter and baked into a cake?

Doing what you want is kind of like that...

There are ways to separate the vocals from the song, but, it's not always a sure thing. Sometimes though, slightly ok results can turn into pretty cool stuff with some processing and/or effects.

[taskrok]

Yeah its not something thats done all that often.
There are plug ins that can do a lot of the work for you but as far as i gather its like eq'ing out apart of the track and the results are mixed.

mess with it though, just google search Vocal removal plug ins or something like that..

[dom]

Have you ever tried un-baking a cake?

Wow, thanks, simply the best explaining counter question i've ever heard regarding this question.


Cheers,
Dom

[Leon Tricker]

The cake baking analogy is pretty much spot-on.

My understanding of how this works is that the plug-in, or the programme you use, assumes that the vocal will be in the centre of the mix.

It then removes frequency information from either side of the mix, and you can EQ/filter what remains to try and isolate the vocal.

In my experience, no plug-in or programme is very successful in doing this.

Is an acapella of the vocal not available for (legal) download somewhere?

[dom]

Yepp, the phase cancellation approach.
But two problems remain: there are other signals also only panned to the center, like solo instruments, bass, etc. - they will also stay after the cancellation.

And the second problem is, that while the vocal may sit in the center, the effect applied to it don't (stereo reverb, delays, chorus etc.) - those will be gone, too.

It's amazing that our ears and brains can concentrate on different acoustic sources like instruments or voices and separate them in a full mix.
This does not work based only on the frequency range but the listening experience and your brain knowing the characteristics of the sound an instrument or a voice can produce. And i guess this is also only the tip of the neuronal iceberg.

Cheers,
Dom

[djsynchro]

Technology marches on...

http://www.elevayta.net/product13.htm

And the guy is charging a ridiculously low price for it!

[master swing]

since they anounced melodyne dna other developers became aware of what could be done, and now they´re putting out their stuff before melodyne. Surely melodyne is not happy about this.

[dom]

I can't imagine that this gives you good results with more complicated tracks.
The demo track is very basic, everything is separated clearly, not only the hard left/center/right panning with nearly no spilling but also the frequencies.
Have you tried the demo with other material? As in all-frequencies-spill-heavy-metal or compressed and fx'ed to death Britney Spears pop?

Or a very simple example in comparison with our brain: Record two people talking at the same time. Your brain can separate them (it listens to the characteristics of the voice far beyond their frequency fingerprint and even takes the context of what was said into account) - try it using this software...

[dom]

since they anounced melodyne dna other developers became aware of what could be done, and now they´re putting out their stuff before melodyne. Surely melodyne is not happy about this.

Na, this visual approach is around for many years. stuff like metasynth (really old) or photosounder (2 years old).

Check this photosounder video:

http://www.youtube.com/watch?v=hUZeE5vM ... r_embedded

[monobeach]

I never did this and am not experienced in phase cancellation, but out of curiosity: what happened to the vocals if you place an instrumental version of the very same tune on a second audio track, insert a utility effect on that and flip the phase?

m

[dom]

This should work, but it would have to be exactly the same mix: sample accurate regarding the timing (i guess two different tape machine recordings would already kill it) and also exactly the same in the dynamics field. Put a compressor with the same settings on the mix including the vocal as well as on the one without you're also doomed again.

[master swing]

since they anounced melodyne dna other developers became aware of what could be done, and now they´re putting out their stuff before melodyne. Surely melodyne is not happy about this.

Na, this visual approach is around for many years. stuff like metasynth (really old) or photosounder (2 years old).

Check this photosounder video:

http://www.youtube.com/watch?v=hUZeE5vM ... r_embedded

Oh, so I guess melodyne has a different approach then, or why would everybody have considered this groundbreaking news?

[weeddigger]

Have you ever tried un-baking a cake?

Wow, thanks, simply the best explaining counter question i've ever heard regarding this question.


Cheers,
Dom

I really wish I could take credit for making that up, and more so, I wish I could actually tell you who I'm quoting!!

Ha...

Maybe it was my alter ego?!?

[Horseface J]

Well, I don't think I made it up...

Then again...


Maybe that other guy, what's his name?!?