Vocal Compression: Save Years of Trial & Error

[@navasaband]

Compression is and will always be essential, but as Andy Wallace always preaches, automating the vocal volume is KEY.

[@kisstherajn]

vocal rider vst?

[@Tsa-SONGS-ABOUT-BEING-a-DAD]

​@@kisstherajnYes or you can draw it in

[@navasaband]

@@kisstherajn vst vocal ride is okay but nothing beats drawing in

[@chefdusse]

@@kisstherajn use the write mode for a starting point then go in and tweak as needed

[@greasybumpkin1661]

Yeah I always go for manual vox riding as a first option, does most of the job most of the time.

[@baytheproducer]

Damn. Never thought of using a waveshaper for vocal clarity. Great video as usual! thanks SA

[@sageaudio]

Thanks for watching!

[@prodbyrog3r278]

Great explanation! Alternatively, for a quick and agressive upfront vocal, try CLA Vocals plugin on 'Face 2 Face Dry', and put the (free!) Xtressor plugin after it on 'vocal beef' setting the input so that it's hitting all of the green lights. This is just for a single lead vocal - for your vocal bus, use a limiter reducing about 3 db to glue all your leads, backing vocals, adlibs together

[@anguzulukebrian3520]

what cla plug in specifically?

[@prodbyrog3r278]

@@anguzulukebrian3520 Is called 'CLA Vocals', by Waves. It seems to be an industry go-to for vocals.

[@benjaminsamama1]

Ha Rogier zit je mijn trucjes te jatten? Hahaha

[@prodbyrog3r278]

​@@benjaminsamama1 Haha ja man!

[@Patrick-ryan-collins]

Very straightforward and helpful. Thankfully now I just need a lifetime of vocal practice and I'm set❤ 🎉 for those key hints at 3:15

[@TokyoSpeirs]

I’d switch the algo to “vocal” with those same settings, it does that soft knee thing you’re talking about while still controlling really abrupt stuff well and also keeping the vocal pretty “alive” feeling without being too squashed. It’s now my favourite go to!

[@sageaudio]

Great tip, thanks for sharing!

[@lingv_26599]

It's time to finish mixing that mixtape...one time

[@sageaudio]

We believe in you!

[@a2bs333]

I love that you simplify....This is how you will outdo your competition. More of this 👏

[@sageaudio]

Thank you!

[@Kenfenit]

The "upfront sound" you guaranteed is not a guarantee at all, because you're not taking into account the source material. With an average recorded vocal, sure you can say that. But giving a guarantee in audio is one of the biggest pitfalls you can fall into

[@roxasblues8225]

When he said "4 to 8 db of reduction" shes talking about relative volumes.

[@ynghendricks8168]

@@roxasblues8225nah you can say what you want, advising someone to compress a „pop“ style vocal 8dB and copy the settings is the wrong information for upcoming engineers

[@finalcut612]

@@ynghendricks8168expecting people to use their brains is just too much, huh?

[@sageaudio]

First and most importantly, Kentendo is a great name. Second, I hear what you're saying! But I stand by this suggestion and could show the rationale behind it if necessary. If you've seen my other videos, you know I never say, "I guarantee xyz," but in this instance, I do, since I mean it! The one variable that needs to be changed is the threshold in order to achieve 5-8dB of attenuation. Otherwise, you can keep the other settings as is or vary them a little bit to get the exact sound you want.

[@SiQuemaCuhh]

@@sageaudio don’t listen to these bedroom mixers lol. People who know more than the bare minimum know what you mean. And if you don’t know Pop vocals are compressed -4 to-8 db sounds ridiculous but in reality it’s the general area to aim for to get that pop vocal. Good video mate thanks.

[@AceDaGodOfficial]

The wave shaping is kind of basically upward comp

[@GloveBunniesVideos]

Great video, as always! Thanks so much. This info is invaluable.

[@sageaudio]

Thanks for watching :)

[@mttlsa686]

Actually, Pro C2 lookahead (but I think it's the same for every plugin) just delays the actual audio that's going to be processed against the sidechain that triggers the gain reduction. It doesn't "literally" analyze the signal before it begins to process like "predicting the future in real time" as many people think.

[@richardbarr7173]

yeah that's correct and helps to explain why increasing lookahead will always increase latency

[@marceloribeirosimoes8959]

It seems that some have a bit of a hard time trying to understand that the material triggers the threshold, and the threshold triggers the compression and the release processes... Once triggered, these processes start and follow the settings...

[@Bittamin]

Yeah and the knee also can change how quickly the compression reaches its full ratio of Gain reduction, too?

[@marceloribeirosimoes8959]

@@Bittamin Sure! The knee is more about how HARD (fast) the compression process will occur. The knee angle means that the compression can happen gradually (raising the action over the material at the very moment the threshold is reached - creating some kind of curve, depending on the compressor you're using) or abruptly (almost like a binary switch on/off). If the compressor gives you the KNEE option, it will regulate the ABOVE threshold material changes. Let's say you set RATIO to 4.0:1. A hard knee angle will hit -4dB as soon as possible and continuously. A softer angle will give you a gradual action over the material allowing a bit more natural dynamics "movements". So, at the end of the day, what do we have? The threshold must be "touched" first, and then we have the ATTACK milliseconds "distance" to start the process. As long as the threshold is been "touched", the compression will be there - and the amount of these changes is measured by the RATIO you've set. And, the last processing part is the RELEASE that, just like the ATTACK, depends on the THRESHOLD to get in action. So every time a material triggers the compression ("touching" the THRESHOLD) the compressor is now processing the material and the full chain is at "alert mode". This material will suffer the process during the time (duration) you've set, from start (ATTACK settings) to end (RELEASE settings). And some compressors give us three more parameters. One, mentioned by you, the KNEE. Other called HOLD - very similar to RELEASE. The third is called LOOKAHEAD. Great for treating unwanted transients. So, when you're adjusting ATTACK, KNEE, and RELEASE, you're setting how the material will be processed WHEN the threshold detects the material - almost like programming how the doors and gates will be locked when a 'motion detector' detects something in the room where it was installed. So, we are talking about the material OVER, ABOVE the threshold. This is different from clippers, for example. A hard clipper (similar to the HARDEST KNEE line) shaves the sound wave, no matter what. But the SOFT clipper starts to change your material BEFORE the threshold, creating a gradual curve of action intensity - treating the material UNDER the threshold line.

[@TheHorseRaddish]

It is surprising for me that compression happens after a signal fell below threshold. Since amplification is calculated via ratio between signal and threshold I thought it would be recalculated on each sample and after a threshold it would become negative or zero. But it looks like real implementation use max value of a window not current value of a sample making the sound smoother not only after threshold but during signal above threshold as well. Interesting!

[@stewie3128]

While this video covers some really important and insightful info, 3:30-5:30 borders on misinformation. The settings you specified can be, in the right circumstances with the right material recorded in a complimentary way, a good *starting point.* But to say that all you need for that particular sound is soft-knee 4:1, 1-4ms lookahead, and a quick attack, is just simply untrue. If the vocalist isn't at a constant distance from the mic, this isn't going to work well. If there's a lot of background noise, this won't help. If the vocalist was eating the mic the whole time, this won't work. There are an infinite number of ways that that particular setting could let your viewers down, so presenting it as "just do this and you're golden" is really quite misleading. You obviously possess the knowledge to educate someone on how to dial in and settle on a compressor setting, which is probably something you do inside your academy. Presenting the information in this manner, though, is going to cause more than a few beginning mixers to spin their wheels on a vocal because they don't have a compressor that is matched to their material. I'd urge you to reframe just those two minutes. Everything else in this video is great, though.

[@sageaudio]

Really appreciate you taking the time to write and share your thoughts, thank you!

[@JiiNxMusic]

Attack, sustain, decay, release. Remember that💎

[@JiiNxMusic]

Also, try mixing upwards and downwards compression together and you can achieve an even greater consistency👌🏾

[@rotkerid462]

saying useless shit won't make you cool.

[@JiiNxMusic]

@@rotkerid462 enjoy your day, champ🙏🏾

[@NZsaltz]

Hi! I thought that the visual compression explanation in this video was nice. However, unless I’m missing something, I think your explanation of waveshaping is incorrect, or at least incomplete. All waves (without DC offset) cross 0 amplitude to oscillate, meaning that you’re introducing distortion to the whole signal, even if you only affect the lower portion. The peaks’ perceived volume do not stay the same. Technically, you only said the peaks stayed the same “level” rather than volume, but that’s easy to miss. This is especially true with a tool like Oxford Inflator, which isn’t linear anywhere, so the peak level can actually increase. Of course, light waveshaping on vocals isn’t a bad idea, and you’re right that using a waveshaper this way may affect quieter portions of the vocals more. However, judging by comments on this video like “The wave shaping is kind of basically upward comp,” I think you’ve failed to communicate what it’s actually doing.

[@richardbarr7173]

this is super interesting and I would include myself in the "sounds like upward comp" crowd! Any tips where I could learn more about waveshaping in this context?

[@morgan0]

2:37 imo there should be a distinction between db linear (what you said and likely what you want) and value linear (the naive method and likely not what you want). but that matters more to devs than users

[@ujin347]

Thank you so much it's inetresting!!

[@sageaudio]

You're very welcome!

[@PolymerJones]

DUDE! Thank you! There are so many crappy compression videos on KZbin. This is the information I needed. Liked and subbed. Are there any plug-ins out there or software that demonstrate what’s happening to the wave form in real time as you are compressing it OR EQing it ?

[@sageaudio]

s(M)exoscope, youlean Loudness Meter 2, and iZotope insight might have the right features you're looking for!

[@PolymerJones]

@@sageaudio I mean like when you put the compressor on and you see the waveform change on the timeline before having to bounce it. Thanks man I signed up for your subscription Sage

[@mahdiyar5734]

This is great keep going ✅

[@sageaudio]

Thanks for watching!

[@stingylizard]

Excellent share.

[@sageaudio]

Thanks for watching!

[@thilofischer3552]

Why use the clean setting and not the vocal setting on Pro Q?

[@sageaudio]

We find it's a bit better suited and sounds nicer for that application, but definitely give both a try yourself and A/B between the two to see what you prefer!

[@3jahel7]

MWaveShaper reminds me alot of Maximus in FL

[@D29.H45]

if there's a sudden loud noise? lets say 10ms and a slow attack at 50ms does the compression still goes full power at the end of the duration or it stops reducing gain and start releasing right when the noise stopped?

[@olanrewajuadagba955]

Thank you.

[@sageaudio]

You're welcome!

[@jeremiahaugust5886]

Definitely going to give waveshaper a try. That greatly peaked my interest!

[@Dave-Rough-Diamond-Dunn]

*Piqued. Not picking on you, it's just so you know next time.

[@sageaudio]

I like "peaked" though :] fits the topic well haha. I'll talk about waveshaping some more in future videos then!

[@Dave-Rough-Diamond-Dunn]

​@@sageaudioTrue! Maybe that whooshing sound was a joke going over my head! 😁

[@francobuzzetti9424]

my cv: 3 years of audio mixing experience also my cv: 15 years of vocal compression experience

[@campar1043]

is wave shaping pretty much upwards compression?

[@sageaudio]

You could definitely say that! It increases the gain of low amplitude signals/transients/waves, while leaving high amplitudes untouched.

[@GafouKafour]

Are 1176 comps are suitable for lead vocal?

[@sageaudio]

Yeah! They work well for rock vocals, but worth trying for other genres in case it fits the sound you're going for.

[@GafouKafour]

@@sageaudio thank you so much and also dear sage audio, could you please tell me how can we use clip gain in fl studio which it isn't as easy as the other DAWs

[@realguitarstuff]

Great video! I always thought you need a slower attack to preserve transients/naturalness and upfrontness. Isn’t the waveshaping emulating a slower attack setting?

[@actosium]

Does BPM or singing vs rapping affect how fast or slow we should engage the compress and release?

[@AceDaGodOfficial]

If its rnb singin then the comp should be a little alower if its rap then the comp should be fast not to fast

[@sageaudio]

They definitely can, it all really depends on the performance and what sort of transient information/dynamic range it ends up having. Always best to use your ears and dial in what sounds best to you on a song-by-song basis.

[@Bleitzofficial]

Hey sage what is a good db to record vocals at? What db should they be peaking at before mixing and after mixing?

[@sageaudio]

Great question, it depends from genre to genre, but somewhere between -10 to -20db if you're in 24bit resolution should be a solid place to start. For mixing, I'll gain-stage during the pre-mix using VU meters and try to keep the average level around 0VU, but after that I'd just recommend relying on your ears to hear what makes the most sense in the context of the mix.

[@WDShorty]

Vocal clarity comes 90% from technique, if they're swallowing their own voice and not using their chest or sinuses or not breathing right, it will sound buried especially in rock music, knowing how to instruct the artist to help them get a better song is important, just tell them when you have the gain on their monitor set that if they can't hear themselves while recording to the song then they're swallowing their voice, then the thing that will truly mess up the vocals and make them sound off is too much reliance on eq, compression or microphones that are coloring the vocals too much, human ears are really good at picking all of these things out as off(subtle parallel compression and a high pass at 50hz is plenty), last make sure they're 6 inches from the mic at the least, avoid too much of the proximity effect, then you can think about adding saturation

[@craetr3014]

how does a compressor amplify a sound if it is attenuating it?

[@paukin9344]

It doesn't, the makeup gain is what increases amplitude after the compressor has done it's thing. If you don't use any makeup gain you will be attenuating and will make the track quieter.

[@sageaudio]

@@paukin9344 hit the nail on the head here!

[@NZsaltz]

To be more clear: because the compressor only attenuates louder parts of the sound, it brings the louder and quieter parts closer together in volume. So when you add makeup gain, as Paukin said, the overall sound is louder than before!

[@NickWeissMusic]

Hard vs Soft knee is the terminology for the info around 2:11.

[@PekkaImmonen]

Instead of attenuation goin up, maybe volume could have gone down? Good stuff anyway! Thanks!

[@sageaudio]

I agree - if I talk about this again in the future I need to make a better graphic haha. Thanks for watching!

[@darkwolfgaming8243]

Can you make a tutorial on how to use tube tech cl1b

[@sageaudio]

We'll definitely consider that in the future, thanks for asking, appreciate you!

[@gangs0846]

Great

[@Rhuggins]

I would pay for a detailed tutorial on Melda waveshaper (and or waveshaping in general). I have been doing this a long time and never thought to use a waveshaper like this. Im serious - would love some info on the transfer curve explanation and techniques

[@sageaudio]

Thanks for watching! I'll look more into it to make sure there isn't something I'm missing, then see what I can go into more depth about in future videos.

[@pimontumba2899]

In Deutschland sagt man „danke“ 👍🏽

[@prod.Kaibit]

Side EQ on a mono signal like a vocal, is crazy..... Unless Pro-C2 makes it stereo or something, side eq on a main vocal is doing completely nothing

[@sageaudio]

I agree! But if you use it after reverb or stereo temporal effects (like how I described in the video) you'll have a side image to amplify due to variations in the left and right channels.

[@andrewrice9383]

Just dropping in to say, I don’t appreciate the overly aggressive title. Thank you.

[@sageaudio]

Vocal Compression, Save Years of Trial and Error...?

[@andrewrice9383]

@@sageaudio yes, it is a strong claim that drives curiosity in someone scrolling through the feed that is out of proportion to what I think is appropriate

[@potatoperson]

@@andrewrice9383what do you want the title to be?

[@andrewrice9383]

@@potatoperson I would probably say something more straightforward like “vocal compression techniques that took me years to figure out“ and then I would remove the “few engineers know this” part of the thumbnail which can play on a sense of exclusivity, which I do not morally agree with.

[@potatoperson]

@@andrewrice9383 lowk good changes 👍

[@mattramen3696]

m’waveshaper *tips fedora*

[@Guilherme-nc5li]

Great tutorial, but the singer sounds Scottish

[@ynghendricks8168]

Not a good video, you are „guaranteeing“ sounds if they take your settings This video shows why learning mixing through these „tips and tricks“ videos will take more time than having a proper teacher. I‘m done with channels like this

[@sageaudio]

Thanks for watching! Yep, I stand behind that! You'll of course need to adjust the threshold to achieve 5-8dB of attenuation, which is how differing performances will be accounted for. Otherwise, all of the other settings can be kept as is, and your vocal will move to the forefront of a mix.