Sitting the vocals in the mix.

[No name]

Well, it sounds excellent. I like the little touch ups though too. The vocal in the beginning that sounds like it's been through a bit of mangling. Sort of like a granulizer (I say this because i'm familiar with the FL granulizer, but it sounds excellent) and then the processed vocal was reversed. I don't know if you like to keep this type of stuff to yourself, but could you share with us all how you went about creating various elements in this track? From processing the vocals and creating the fx? I ask a lot!!! Which of course you might be busy anyways so it's no big deal if you can't or don't wanna, but it would be enlightening to see what you've done with it.

I'm happy to share my techniques. You're right about the mangled vocal - I took the lasst phrase of the first verse, send it through KT Granulator (love that plugin!) with lots of feedback, and then reversed it.

Were there any other sounds in particular you wanted me to explain? I only ask because it'd take me all day to explain everything! Even relatively "simple" things can get pretty involved.

For example, there's a background pad in the introduction that sounds pretty basic, but it's back there to hint at a slightly more complex harmony than explicitly stated by the bassline on the tonic. The three chords in the pad move gradually from an "open" stacked fourth (B-E-A), via the flat 6th in first inversion (one of my favourites - B-D-G), before finally arriving at the root tonic (B-D-F#) - the effect is of gradually resolving to the tonic. At which point the violin also arrives at the tonic, and the vocal slice starts to emerge out of the texture. At first the vocal slice sounds like percussion due the the strong sillibance (and the similarity to the established hihats), but just as we start to recognise the sound as a voice, we hear the first word of the song - "love". The synth pad also makes a couple of reappearances to subtly add some harmonic diversity, but it's too subtle to pick out on it's own.

Or the synth bass in the introduction. It's one MIDI sequence simultaneously sent to two synth parts. The MIDI data has all random velocities (implemented by a Logical Editor preset I made myself in Cubase). One synth (panned left in the final version) has a resonant low pass filter cutoff controlled by velocity, and then massively distorted. This results in a harmonically rich sound that constantly changes spectral balance with every note (sequenced 16ths mostly) without the inconsistent volume changes that commonly result from "regular" filter cutoff. Maybe next time I'll try a phasor. the other synth line was just a straight electric bass, but processed with DFX Buffer Override, with the LFOs set to random drift, and the buffer size tuned so that the effect is similar to oscillator sync... if the oscillator was an electric bass, and the sync signal was drifting all over the place.

And that's not even half of what's going on in the introduction!

Ohhh, got an idea. A thread should be creating where people will put together their music, and then they will explain the process that led up to the creation of the song, it would be an excellent reference for others, as well as a chance to grab some inspiration. Maybe i'll post one in the music cafe or something, but i'm not sure if anyone will go with it or not, but we'll see.

I've actually thought that a cool idea would be to have someone work on a song, but upload the latest version at the end of each day (or each hour, for me ) so everyone could hear the progress and the process as the song develops.

-Kim.

Very interesting. I'm even more interested now! I think it's interesting how you use your chords and all, as i'm really just beginning to have some depth in understanding my music theory, and it's coming along pretty well. I used to be like a wild monkey pounding piano keys until a decent string of notes come out, and then I had no clue wht they even were . Clever little trick you used for the bass in there too at the intro part. Fantastic production. You've inspired me more to pay attention to detail in my music, and take the time to process each instrument just the way I want it.

All these years and it never has really deeply occured to me that it is best to have the vision, and then you use these skills you've learned to put that vision into reality. Music is such a lovely thing. This is why I tend to shy more and more away from doing hip hop anymore, because there really isn't as much room to play because folks want it a certain way, there is no real way to experiment. That is all philosophy though, so back to the topic...





You used things that i'm not totally sure of what they are. Such as the concept of a "Logic Editor".

I know one thing, all the things you've implemented, my PC would have a stroke trying to keep up with all the different things going on. As a matter of fact, I think I would have a stroke as well! Very enlightening! I'd particularly like to understand what you did with all the different instruments in the song, EQ and sound spectrum wise. what you cut, how you fitted it all into there. There is a lot going on, but nothing seems to really overlap.

By the way, is that you singing? Lovely vocals, very dramatic and emotional.

[Kim Lajoie]

Very interesting. I'm even more interested now! I think it's interesting how you use your chords and all, as i'm really just beginning to have some depth in understanding my music theory, and it's coming along pretty well. I used to be like a wild monkey pounding piano keys until a decent string of notes come out, and then I had no clue wht they even were .

Ha, you'll love the chorus chords then. It's based on Bm-D-F#m-Bm, with the vocals starting two bars in, on the D. I used that as a two-step transition, and overlapping the bookends (both Bm) on the double choruses (and why I did that is another essay in itself ). In practice though, I extended the chords to something like Bm7-C#9-Dj7-E9-F#m7-Bm7... I can't quite remember. It's actually very close to the 2nd time chorus from Gwen Stefani's monster hit "What you waiting for?" from the first album "Love Angel Music Baby". No such thing as originality.

Clever little trick you used for the bass in there too at the intro part. Fantastic production. You've inspired me more to pay attention to detail in my music, and take the time to process each instrument just the way I want it.

That's one of the great things about the computer studio. We have the technology and the freedom to realise exactly the sounds we imagine.

All these years and it never has really deeply occured to me that it is best to have the vision, and then you use these skills you've learned to put that vision into reality. Music is such a lovely thing. This is why I tend to shy more and more away from doing hip hop anymore, because there really isn't as much room to play because folks want it a certain way, there is no real way to experiment. That is all philosophy though, so back to the topic...

That's the way I see it. I imagine the music first, and most of the composition process is getting the sounds out of my head and into the physical world.

Nothing wrong with hip hop, by the way. Every genre's got room to bend and explore. The problem comes when you're doing work for a client, and they're calling the creative shots. Of course, that's not restricted to hip hop. I do a lot of film work, and I get ... uh .. "all kinds" of directors.

You used things that i'm not totally sure of what they are. Such as the concept of a "Logic Editor".

The Logical Editor is one of the MIDI editing tools in Cubase SX. Rather than editing individual notes, it's more like an advanced rule-based "search and replace" or "search and transform". The settings I used for the mangled bass was a set of rules that essentially said "Get all notes, and replace their velocity with a random number".

I know one thing, all the things you've implemented, my PC would have a stroke trying to keep up with all the different things going on. As a matter of fact, I think I would have a stroke as well! Very enlightening! I'd particularly like to understand what you did with all the different instruments in the song, EQ and sound spectrum wise. what you cut, how you fitted it all into there. There is a lot going on, but nothing seems to really overlap.

Actually, I don't use a lot of EQ. Most of the spectral placement is is sound choice and composition. By far the most common EQ I use is a hipass filter.

To give you an idea of the size of this project, if I recall correctly, there were:
18 SampleTank2 channels
5 main vocal tracks (intro, verse1+2, chorus, chorus harmony1+2)
10 background vocal yelling in the chorus
1 electric guitar (with Amplitube2)
4 printed (bounced) and imported effects (reverse snareclap, feedback build up, violin solo, reverse reverb)
1 Reverb send (CSR)

Most of those channels would have had one or two effects plugins. I'd consider that a medium-sized song.

A big project would be on the order of 48 SampleTank2 channels, 20 or 30 vocal tracks, another 10 or 15 guitars, and another 20 tracks of printed and imported effects. Those projects usually take longer than a week.

By the way, is that you singing? Lovely vocals, very dramatic and emotional.

Nope, that's my talented sister.

-Kim.

[Martin E]

I took a listen to your song, but I'm still up in Vancouver using the in-laws computer with speakers that have no low end at all and that I'm not familiar with, so for me to give too many specific critical comments on your mix would not be too helpful since I wouldn't know if it was a problem with your mix or these speakers.


Steve

Thanks for the comments, Steve. And I'll have a look at the 500 Hz area. I did cut the vocals at 500 Hz as you suggested in another post. Which has a very positive effect on the vocals. As for bringing out the vocals more I must say I'm not a big fan of very loud vocals. Call me old-fashioned But maybe a touch will do.

The song itself was more or less an attempt to create a type of song that would have the feel of the STAX soul sound. It's not a style I work in a lot. But I thought it would be fun.

If you have time to listen on better speakers I'd appreciate it.

[chilln]

BUMP! Excellent stuff, guys!

[chilln]

Stickie??!! Please?

[ahjteam]

The original question was, "how to make the vocals sit on top of the mix". The three things are:

* Make the vocals louder
* Make everything else quieter
* Use reverbs and delays

First of all, I am more of a live-engineer myself, and these might sound a bit cruel to studioguys, but they are pretty much how it works without ruining the overall sound. What you need is:

* Spectrum Analyzer of your choise
* A parametric EQ of your choise (you can also use the built-in ones if the host has one, they usually aren't bad at all)
* Optional: A compressor with sidechain (use low ratio), if your host supports this feature
* Optional: reverb and tap delay of your choise

Step 1: Group the vocals
Step 2: Open the spectrum analyzer and look where are the main peaks in the vocals (my guesses are around 500, 1k, 2k and 3k), take a screenshot, open it in your favourite image processing software (even MS paint will do for this purpose) and pick atleast three(3) of the highest points (unless they sound like doodoo, remove them from the vocaltracks).
Step 3: Put everything else in a group and remove 1-5db with pretty tight Q from those frequencies. If the peak was high in the vocals, remove less and vice versa.
Step 4: Sidechain compress the instrument-group with REALLY low ratio (like 1:1.2), slowish attack and release and low threshold so that even the slightest vocalparts makes the compressor react.

Of course, this spectrum analyzer thingy usually doesn't work in live situations, because in 9/10 cases the vocalist usually gets really frustrated atleast within few minutes if (s)he has to sing alone without the music. Also the feedback is usually a problem and you won't have EQ's in groups on most (atleast reasonably priced) consoles. Also the lack of sidechain compressors (or atleast they usually aren't connected, even if they have the feature), is the reason why you just have to use your ears and use reverbs and delays to make the vocals come out clear.

I prefer "Hall" type of reverbs that are on most hardware reverbs. And usually "Hall 2" preset, which sounds pretty large, modified so that it has 2 second decay and maximum hi-ratio (=the least amount of treble possible).

For delay I prefer TC-electronics' D-two or any tap tempo delays. Long delays (like 600-1200ms, aligned to the tempo of kick/snare) with a lot of feedback (like 30-60%) but only 2-3 echoes and low volume. I rather use the delay to highlight some parts, rather than use constantly in the background, like some guys do.

So, Step 5: Add slight reverb where predelay is somewhere around 40-300ms. This is really much depending on the vocal style, if its moody, use longer, if its really (like occasional shouts and growls), use shorter.

And especially in live situations only with cutting (not boosting) EQ on vocals works miracles, but same thing is in the studiorecording. Also if you have to make something louder, try to make the signal-level to 0 and rather use the output slider than the input gain to minimize the risk of feedback live and hiss in studio.

[chilln]

bump