Can anyone help me with volume automation options?

[purplepeople]

Back in the day, I would record just a couple simple layers of vocals so it was easy to match everything up.

Recently, I've been recording tons of layers, and sometimes vocals line by line. This is creating sessions with 100-200 tracks.

This is making it very time consuming to properly match up the volume. Not only within a verse, but from verse to verse.

It can take me a full week just to do one song. I'd say 99% of my mixing time is spent on volume matching.

Are there plug-ins or options to easily match up the volume of all clips?

Like if I wanted all of my main vocals to match. All of my backups to match. All of my pans to match. All of my adlibs to match. etc.... from verse to verse.

If so, what is the process?

I mix in the newest version of Adobe Audition.

Thanks so much for your input/help.

Ps. I have heard of "Vocal Rider," but I haven't tried it. If I created 5-6 different vocal busses could I in theory place this on each one? Or would that take too much processing power?
I like to keep my breaths at 1%. Wouldn't the program in theory raise them decibels? Or would you just set it up to only reduce volume instead of raising it?

 

[Papanate]

Off the top of my head a Normalizer Plugin would do it all at once (you could batch process it) - but not that you need to hear this - but why aren’t all you vocals the same? Do you tear down you system all the time?

 

[purplepeople]

Off the top of my head a Normalizer Plugin would do it all at once (you could batch process it) - but not that you need to hear this - but why aren’t all you vocals the same? Do you tear down you system all the time?

That would just normalize the overall volume of the entire clip right? It wouldn't treat each word of a clip?
The vocals aren't the same volume because the performance is very aggressive rap. Meaning some words are much louder than others. Lots of punch ins.
I also use multiple styles/voices/rhyme patterns within the same verse. It's very dynamic/cartoony intentionally to make for a very exciting/hype sound.
Very 3-D'ish with tons of panning. Nightmare to mix, but the end product comes out cool. I'm looking for solutions of turning 1-2 week long projects into a matter of hours.
The way I've done it for years is making a full mixdown of everything from the same layer for all of the verses. Then I manually go in and adjust the high and low parts of each line to match the last. Then I do it for the other layers. Which I'd keep doing if I only had 2-3 layers, but I have more like 20. It's way too time intensive.

 

[Slouching Raymond]

purplepeople has a point there, but normalizing each vocal track would give you a good starting point.
I routinely use the Cubase 'normalize' feature, which came as standard.

 

[Papanate]

That would just normalize the overall volume of the entire clip right? It wouldn't treat each word of a clip?

It would raise each clip to a nominal level - Say -10db or so - you would apply to each of the 20 tracks of vocal comps you have.

The vocals aren't the same volume because the performance is very aggressive rap. Meaning some words are much louder than others. Lots of punch ins.
I also use multiple styles/voices/rhyme patterns within the same verse. It's very dynamic/cartoony intentionally to make for a very exciting/hype sound.

That’s sounds pretty cool -

Very 3-D'ish with tons of panning. Nightmare to mix, but the end product comes out cool. I'm looking for solutions of turning 1-2 week long projects into a matter of hours.
The way I've done it for years is making a full mixdown of everything from the same layer for all of the verses. Then I manually go in and adjust the high and low parts of each line to match the last. Then I do it for the other layers. Which I'd keep doing if I only had 2-3 layers, but I have more like 20. It's way too time intensive.

I’ll look around - but on a word by word basis I don’t think you can get there - GainMatch by LetItMix might get you there - but on a word by word basis I think you would need to do it the way you’re doing it now -

 

[purplepeople]

It would raise each clip to a nominal level - Say -10db or so - you would apply to each of the 20 tracks of vocal comps you have.

That’s sounds pretty cool -


I’ll look around - but on a word by word basis I don’t think you can get there - GainMatch by LetItMix might get you there - but on a word by word basis I think you would need to do it the way you’re doing it now -

Anytime I've used normalization, it simply takes a measure of the loudest part of the track and then matches from there. The problem with my vocals are the large peaks. Like the peaks in some places are 6 db louder than the rest of the clip. I'm trying to find a quick/easy way to reduce those peaks without having to manually adjust each one.

 

[rob aylestone]

This doesn't work if you are chopping songs up into phrases and then comping. I'm surprised though that it's causing such grief. If you have loads of takes, then are you not comping traditionally? In cubase, that I'm using, I colour code every take - so a dark green is a good one, light green nearly, down to solid red rubbish - then when the entire song displayed on the screen has all green on the different tracks, I save it, delete everything not green. I'm then left with a complete lead vocal spread across maybe ten tracks? I then bring all the darkest green takes up to new empty vocal track. I might be left auditioning the best ones a few times, but then I do the same with other tracks.

Level wise one phrase might be loud and the next soft and that is correct - so normalising does NOT work - you still need fader adjustments, but the way I do it - you have the entire song with the best bits that will be at the right rough level, so once the track is comped - fader moves can be done simply and quickly. if you have not comped properly, then it will take for eternity. Normalising a chopped up song sounds simply dreadful - some phrases are not meant to be loud.

 

[Folkcafe]

If considering Vocal Rider, here is a recent interesting video on setting it up to utilize its built-in write automation.

 

[Papanate]

Anytime I've used normalization, it simply takes a measure of the loudest part of the track and then matches from there. The problem with my vocals are the large peaks. Like the peaks in some places are 6 db louder than the rest of the clip. I'm trying to find a quick/easy way to reduce those peaks without having to manually adjust each one.

I didn't fully understand what you wanted - normilzation would make a muck filled track - Waves Vocal Rider (only $29) would do want you want - watch @Folkcafe video above for the low down.

 

[purplepeople]

If considering Vocal Rider, here is a recent interesting video on setting it up to utilize its built-in write automation.

Thanks. His process makes sense, to simply go down in volume instead of up and down. This way breaths can be kept very low without it increasing them.
Instead it just reduces the peaks, which is what I'm looking to do. This way the vocal is evened out before ever applying a compressor or anything.
I presume it would give you a much better mix since the peaks wouldn't be smashed. Also less compression could be used.
I've never tried a "write" function. Does Audition have that capability?
Can it be added to a bus?

 

[Papanate]

Thanks. His process makes sense, to simply go down in volume instead of up and down. This way breaths can be kept very low without it increasing them.
Instead it just reduces the peaks, which is what I'm looking to do. This way the vocal is evened out before ever applying a compressor or anything.
I presume it would give you a much better mix since the peaks wouldn't be smashed. Also less compression could be used.
I've never tried a "write" function. Does Audition have that capability?

Vocal Rider is not compression - it is just like riding a fader - only it goes both ways - and it’s $29 at Waves.

 

[purplepeople]

Vocal Rider is not compression - it is just like riding a fader - only it goes both ways - and it’s $29 at Waves.

Vocal Rider is not compression - it is just like riding a fader - only it goes both ways - and it’s $29 at Waves.

Right, but using something like this would enable me to use less compression on my vocals. Which is what I'd prefer to do.
I've had it for years, but I was using it for a different function. Never tried the write feature. I'm going to experiment with it. Hopefully it works on a
bus because it would be incredibly tedious to set it up on 100 different vocal tracks.

 

[Folkcafe]

Right, but using something like this would enable me to use less compression on my vocals. Which is what I'd prefer to do.
I've had it for years, but I was using it for a different function. Never tried the write feature. I'm going to experiment with it. Hopefully it works on a
bus because it would be incredibly tedious to set it up on 100 different vocal tracks.

One of the important points he makes in that video is that by working this way, levels are within a consistent range when they hit the compressor.

 

[drtechno]

The problem with my vocals are the large peaks. Like the peaks in some places are 6 db louder than the rest of the clip. I'm trying to find a quick/easy way to reduce those peaks without having to manually adjust each one.

Look ahead limiter plugin then afterwards a compressor plugin. Since you didn't track it with a limiter going in, which is the correct thing to do instead of a compressor when recording into an ADC. That is one of the ways the interfaces are built wrong, but that is a different subject.

 

[DM60]

I have used vocal rider. It does work and it allows you to record the automation so that the plugin can reuse it. It is rather lightweight and can be used on each track. You can also use it on the final track (like a group) and run all of your comps to the group header track.

Now how good it works, I would have a had time to say as I usually do it either manually like folkcafe presented, or use a compressor to level it out. I only manually edit the wave (I use Ableton so it is a non-destructive function) to tame the peaks so I can be less aggressive with the compressor. I am also not a sensitive to sound as many others on this board are.

 

[pj11]

I'll throw waves trackslammer out there. I've always had good results using it, especially with multi part harmonies. Makes it so much easier to blend and tighten up.

 

[TimOD]

To me, the process would be to consider that perhaps you're going Too Far with all those vocals. One hundred to two hundred tracks of vocals? That seems excessive, to say the least.

 

[rob aylestone]

I have a friend who does this routinely. We both use cubase and we send stuff to each other to work on. He is a guitarist, i might get 100 guitar tracks, each little twangy bit on a new track. A three minute track with two seconds of guitar. Many even have their own eq for that short bit. I just cannot work with that many tracks, so my first job is to find the ones with the same eq and effects and move them to one new track ‘guitar with fx1’. Then i do the same on the others. 100 guitar tracks down to maybe ten? I can work with that. He also has hundreds of plugins active. For me, i hate hearing something and not being able to see it. Scrolling down through so many tracks is awful. Volume matching is then also way, way easier and no automatic levelling needed.

 

[jimmys69]

Looks like Audition has 'Clip Gain'. I would start with visually chopping up the vocal and levelling out the obvious before getting into automation. Automation can get messy really fast. And if the levels are really out of whack, a compressor/limiter isn't likely going to work the way you are wanting.

I still have not tried Vocal Rider. Sounds like it could help, but I always just go with clip gain adjustments for the most part.

Some Videos that show a few ways to skin purplepeople's cat:

adobe audition clip gain video - Google Suche

www.google.com

adobe audition clip gain video - Google Suche

www.google.com

 

[LazerBeakShiek]

""Limiting is kind of necessary. That way you can play or sing and the volume stays the same. It becomes mixable.

I dont place anything in front of the input on the preamp. I use the effects loop and limit it post. Just before the power amp. same with voice.

Vocals work great with a cheap limiter. Grab an Omni pattern microphone. Then something like a FMR RNA and PBC-6A team, will do a nice job. That set, sent to a tube power amp and then PC interfaced/recorded into the DAW on a Suhr reactive load. Reactive load, taking place ofa 8 ohm PA speaker. The other PA speaker can be ended with a dummy load 8 ohm to keep the work area silent.

A compressor is just a magnifying glass to augment the sound parts you like...and make it growl. The limiter evens it so every note can be heard.

A terrible sound/video clip I made on my cel, shows the bass stays the same volume, high strings to low. You know that that aint going to happen without a limiter. The low strings are much louder normally and clip when you slap. Notice the LED lights on the Micro Limiter, how the limiter works harder as the notes get lower, keeping the volume even. Amazing. Pound on it all you want. Won't matter.



My bass rig is an all tube VBA 400 and ADA MB-1 w/ respective cabinets.
Fender Jazz Bass V (USA 5 string low B)
The limiter I use is an 80's Alesis Micro Limiter (uncommonly good)
*That is my heater on in the background* ""

yeah..

I practice hard limiting, always.