vocal recording and processing

[shure_1302]

Hi !
(please move the thread if it should be somwhere else).

I need very simple advice:
how should I process vocal - should it be the chain mouth-mic-preamp (with initial procesing - i.e. DBX 286A)-audio interface-PC or better do it mouth-mic-preamp in audio interface-PC and deal with everything via software? I guess it could be done both ways, but what would be pros and cons?

Another thing - I have looked in books and in the net - is it that right sequence of effects for processing vocal?

compressor/limiter-noise gate-equalizer/deesser-reverb

I did not acually find the simple scheme with explanations why it should be like this or like that.

I would be happy to hear the opinion of the experienced person. Or just a link to a good book/article. Thanks.

[AfaraWayland]

Originally Posted by shure_1302

Hi !
(please move the thread if it should be somwhere else).

I need very simple advice:
how should I process vocal - should it be the chain mouth-mic-preamp (with initial procesing - i.e. DBX 286A)-audio interface-PC or better do it mouth-mic-preamp in audio interface-PC and deal with everything via software? I guess it could be done both ways, but what would be pros and cons?

I have two 286s, but I mainly use them on kick and bass during live performances, IOW, i never really use them while recording.

The pros are: you can achieve a sound you like
The cons are: you can achieve a sound you don't like.

Most will tell you there are no rules or presets to follow hear, just experimentation.

The thing about outboard processing is that once you've recorded it, you can never remove the processing, only alter it further.

Most of my EQing, and compressing/deessing are ITB. I usually stick with mic and pre amp to A/D to interface, adding EQ and COmp later if necessary.

What it comes down to is your interpretation of how it sounds.

If you think you hear a better sound one way, use that way as opposed to the other.

The same goes with vocal chain processing priorities. There are things I do, for example, EQ-COMP/De-ess-EQ-effect, but it changes in different situations. The ultimate goal is to capture exactly what you want to hear, and not HAVE to process the signal at all.

[Charlie_M]

Like a lotta stuff man I think you gotta try it both ways and see what you like best. I'm prety new to recording and am jsut geting a rig together. I recently added a mic pre with compressor and eq onboard. I tried that and decided I liked the compressor and EQ in protools better. Not that they are superior in capability - they might be I don't know but I kind of doubt it. What I liked about doing the compression and EQ in software is:

1. I can tweak it and sleep on it and then go back to it later to see if I still like it. If I hear it and say "ah geez what was I thinking ?" then at least I can play with it some more. Probably the need for that becomes less as you gain experience but for the moment I prefer doing it in software.

2. Other thing is I find the EQ and compressor graphics that protools provides to be very intuitive. I'm sure that's also true for other software packages. I'm just a beginer in this and don't know what the heck I'm doing and even I can converge on settings I like pretty quickly that way. Maybe I'll reach that point with cutoff frequency and rolloff knobs on analog gear at some point but... it's not high on my list of developmental priorities so, for me personally, I may not ever.

[richiebee]

Good engineers can apply signal processing going in, because they know exactly what they're going for ahead of time. The rest of us normally apply all signal processing afterwards. It's safer, because we can always undo what we don't like without ruining a take. I would recommend recording clean and dry at least until you really know the exact settings that work for you. If you're recording someone else, you really probably don't know the exact settings that work!

You can split processing into two basic sections - inserts and sends. Although sequencers like Cubase often allow you to use effects like reverb as insert effects by having a wet/dry balance, they are essentially sends. Insert effects CHANGE the signal, send effects ADD to the signal. Compressors, limiters (limiters are compressors!), EQ's, noise gates are usually used as insert effects (de-essers are basically narrow band compressors... could be considered EQ/compressor hybrid!). Reverbs and delays are usually used as send effects. You can get interesting effects by going against this basic set up, but for the standard, that's where its at.

Signal processors by their very nature will affect the signal going into the next signal processor but also what has been processed before. So, its a game of compensation... set up a signal processor perfectly, and as soon as you apply the next signal processor, your first one is no longer perfect! If you apply the EQ and THEN the compressor, the EQ will have an effect on what gets compressed. If you apply the compressor THEN the EQ, the compressor will influence the signal being EQ'ed. As long as you realize this, the order becomes less important to get a result, but more important to the ease of how you get that result.

If it was me, I'd generally go

noise gate/compressor/EQ/limiter

If you use each signal processor in moderation, this order should minimize the amount by which you use one processor to do the job of another.

If you're not into moderation, this order may well change!

Remember that a limiter is a compressor, usually with fairly heavy settings. A de-esser is basically a compressor/EQ hybrid.

If you need to use a noise gate on vocals for essential (rather than creative) purposes, maybe your set up needs review.


R.

[shure_1302]

Thank you everybody!

The ultimate goal is to capture exactly what you want to hear, and not HAVE to process the signal at all.

I guess it is very reasonable.

To Charlie_M - yea, I am not going to be a pro in recording as well, but it is good to know basics anyway.

To richiebee - thanks for all teh explanations - do you know a good book with theory and audio examples as well?

But still - if you do not know what to seek it is very hard you find anything, right? What properties distinguish good recorded track from the bad? Let's say I made to takes using two different ways of recording/processing and I could hear the difference (even I doubt it. But which will be better? It is hard for me to judge. The answer "decide for yourself" is good one, but usually when pro would listen to them he could easily pipoint the pitfalls and mistakes and explain why is so and why one take is better than the other and "here the compression is too high..." "here you have to little mid frequencies" - and it become obvious. But I would realized it only when somebodt shows me it.

Definetely I fo not consider here popping are touching the mic, ssss or any other obvious mistakes.

[AfaraWayland]

But still - if you do not know what to seek it is very hard you find anything, right? What properties distinguish good recorded track from the bad?

Try comparing your sounds to the sounds on your favorite records.

Who is to say that there is anything you should be seeking...we are not you. Nor can we say that one or a few ways satisfies the whole. Many different ways satisfy many different people in many different ways.

If you goal is to please as many people outside of yourself as possible, ther are objective ways of getting generally pleasing sounds. But...you will not please everyone. So, the object is to strive and please as many people as possible (in your intended audience).

Different audiences like different things. Figure out what audience you are trying to please, then make your stuff sound comparable to those things.

For example, if you are going for radio-like commercial alt rock, then listen to that alt rock, and try to emulate the sound of it.

When I say this, I am meaning not just the whole perspective, but the individual elements as well.

The answer "decide for yourself" is good one, but usually when pro would listen to them he could easily pipoint the pitfalls and mistakes and explain why is so and why one take is better than the other and "here the compression is too high..." "here you have to little mid frequencies" - and it become obvious. But I would realized it only when somebodt shows me it.

Unfortunately for all of us, we each have to teach ourselves those things. One thing that helped me was a course in critical listening.

Listening occurs on many levels. One needs to focus very hard on what they listen to in order to start seeing the depth involved.

Take the frequency spectrum for example.

A good start for it would be to invest in a copy of "Golden Ears" CD set.

another and less expensive way would be to put a pink noise generator through a parametric EQ on a track, set a narrow Q for eq bandwidth, then boost the band about 6 to 12 dB. Move the band around from low to high.

You will notice the character of the pink noise change as you sweep the EQ.

Eventually with practice, you will know the properties of the different octaves of the frequency spectrum, and then, identifying and solving EQ problems with a track could come easier.

The same sort of experiment goes for compression, though I would suggest reading up on the theory of it first, before testing/experimenting with Envelopes, attack, release times, comp ratios, and so on.

The reason why the "pros" can identify the things you speak of is because they have learned (taught themselves) what to listen for.

There are numerous books on topics like this, surely a neighborhood bookstore would carry at least one in the music section. Otherwise you can find them on the web.

But realize this, that book smarts are greatly different than practical application and material absorption in the physical world.

[brandondrury]

The first thread kind of gives me the vibe that you are supposed to process vocals. I don't really feel like I'm supposed to do anything. I only EQ vocals when I change my mind about a vocal sound. In other words, maybe I thought I liked a certain mic during tracking but during mixing I find that I'm not happy with the tone.

Actually this illustrates a huge point. Your tracking brain is way different than your mixing brain. There are a billion reasons for this, but I think I would always prefer to have a natural sound that I can always go back to after I decide that I don't need to process the vocals nearly as much as I already have.

I recently added a mic pre with compressor and eq onboard. I tried that and decided I liked the compressor and EQ in protools better.

Most big boys prefer the real thing, but they aren't using ART EQ and compression. They'll be using a $2k EQ (if any EQ at all) and a $2k compressor. On top of that, there are probably a handful of $4k EQs and compressors that they didn't choose because they weren't the right tools for the job.

I've tried a few inexpensive outboard compressors and I like the Waves Rcomp better. With that said, the plugins in the UAD-1 are supposed to be much better than the Waves plugins (so I hear).

Other thing is I find the EQ and compressor graphics that protools provides to be very intuitive.

I'm not sure if this is all that important. It's easy to get sucked in by the eye balls. The hardest part about recording on a computer is not listening to your eyes and listening with your ears.

If you need to use a noise gate on vocals for essential (rather than creative) purposes, maybe your set up needs review.

I can't recall every using a gate on vocals...except as a reverb / delay expander (where I only want the reverb to kick in on loud parts). That is a fairly advanced mixing trick and not really something I use very often.

What properties distinguish good recorded track from the bad?
The same thing that determines good cooking from bad....if the people eating it like it, it's good. If they don't, it's bad. That's it! Every rule has been broken a million times. The only rule I know for sure is to make sure that nothing hurts. If anything is so harsh that it hurts your ears, something is wrong.

but usually when pro would listen to them he could easily pipoint the pitfalls and mistakes and explain why is so and why one take is better than the other

Sure. But then the next pro will have a completely different opinion on the same track. There is no standard for a specific instrument / voice. However, there is a standard for an overal mix. If the song works, the mix worked. If the song doesn't work, maybe the mix sucked...maybe something else sucked.

Try comparing your sounds to the sounds on your favorite records.

If you do this, put in about 10 of your favorite records. Just one or two won't do you any good. You'll think your snare should always sound like X and this is the path to disaster. After hearing 10 snare drums, you'll get a better idea of what your monitors are telling you and what is expected of you.

The reason why the "pros" can identify the things you speak of is because they have learned (taught themselves) what to listen for.

In addition to that, they have their own tastes which happen to be dramatically different from one engineer to another.

I want to say that engineering is NEVER seperated from the music. I'm of the opinion that very very very few tones are saught after until someone has had a socially significant record with that tone. People copied Eddie Van Halen's tone of the first Van Halen record because the entire record is awesome and completely innovative. The music still inspires people who don't give a damn about guitar sounds.

If you think for one second that you do great engineering on a piss poor record, you are going to find yourself kicking yourself after a few years (like I did). You are making music. If you are ONLY engineering records, you are at the mercy of the person you are recording...good luck!!

Brandon

[AfaraWayland]

Originally Posted by brandondrury

People copied Eddie Van Halen's tone of the first Van Halen record because the entire record is awesome and completely innovative.

right on,

and to illustrate another point,

Van Halen played in dive bars for ten years before breaking out.

word.