Fixing the voice to obtain a uniform level (TIPS PLEASE)

[demoniobox]

Hi long time since my last posts here,im back on the recording area now !
I bought M-Audio's mobilepre usb and i think its turning great for me !
please read this topic :
https://homerecording.com/bbs/showthread.php?goto=newpost&t=217294
Im recording just voices now and i don't know if this topic should be on Recording or Mixing,but i think here is the right forum.-
Im newbie at some techniques and i want to learn how to do this :
Im recording vocals on Cool Edit Pro,and sometimes is some part of some songs the vocals are too loud,and the other parts are real soft,you know its like Mother Mother (Tracy Bonham) its a soft song and then the screaming comes...well.that's my problem,this parts are louder than the soft parts
what tool or software,or maybe a filter on my old cool edit do you suggest me to obtain a uniform level on the whole voice track ?
So i can mix the whole voice track into the music easily and not almost word by word or "scream by scream" cutting and reducing the volume...that's a real pain !!!!!!!
Thanx i hope you help me !

i think maybe i need to compress the voice track ?
just guessing !

 

[SouthSIDE Glen]

Im newbie at some techniques and i want to learn how to do this :
Im recording vocals on Cool Edit Pro,and sometimes is some part of some songs the vocals are too loud,and the other parts are real soft,you know its like Mother Mother (Tracy Bonham) its a soft song and then the screaming comes...well.that's my problem,this parts are louder than the soft parts

Before you start throwing technology at it I'd suggest throwing some technique at it first. Work on you mic technique when singing, learn huw to pull away from the mic when screaming and get up close to it when whispering and so forth. Watch all the great vocalists and they "work the mic" like it's another instrument. All that movement is not just for dramatics.

AFTER you have the technique down failry well, the amount of technical compression you'll need (hardware or software, if any) will be reduced and the quality of overall sound will be improved.

G.

 

[demoniobox]

thanx for your answer,yes,i know that part,but there are some slightly variation in volume when the screams come in,but yes,that's a good tip and i already know that
that's why im asking for the software help,the whispering sounds real good and the screams too ( im not saturating, ) but now i think it's software time
cause the differences are not that big thanx to the technique you mentioned above
thanx !

 

[SouthSIDE Glen]

thanx for your answer,yes,i know that part,but there are some slightly variation in volume when the screams come in,but yes,that's a good tip and i already know that
that's why im asking for the software help,the whispering sounds real good and the screams too ( im not saturating, ) but now i think it's software time
cause the differences are not that big thanx to the technique you mentioned above
thanx !

In that case, you can the apply some light compression with a compressor. But I'd recommend applying it to the track during mixing, not during tracking; i.e. record your vocals in their natural dynamic glory, and then apply compression as desired to the already-recorded track. The idea here is not to pidgeon-hole yourself or reduce your options by making assumptions on how the track will sound the best before you've tried mixing it with the other tracks. By recording au natural (more or less), you can always go back to the original recording and try something different in case what you figured you'd need doesn't quite work for you (this happens often.) But if you compress on the way in, you're stuck with that that compression as a starting point and can never undo it without artifacting the sound more.

There is an entire army of free compressor plugs available for CEP. Go to kvraudio.com and search for compressors. Any VST or DX plug will work for you. You might want to start by looking at the "Blockfish" compressor by Digitalfishphones, though there are many others that'll work well also.

G.

 

[pod4477]

ya man i had the same problem. im using cool edit pro too. I used a compressor for some parts because a few parts were louder in the verse than all the other parts.

 

[demoniobox]

thanx !
yes,i think compression is the way to go!
well,what i do to "reverse" in case that i dont like the compression
i always duplicate the track and experiment with it
but i always keep the original muted ! and i experiment with the clones
thanx

 

[Joseph Hanna]

I think I'd work with volume automation first using a meter and picking a random "average" level. Say -8 to start. From there I try to seat the vocal track (that is staying with the average level but raising or lowering it in it's entirety) ignoring the times it's to much or too little for now. Once you find an average level that sits okay I'd begin the process of automation. Easy enough to automate in segments of whole verse's and chorus's or by whole lines or once you get good at it by individule words.

This of course requires a good meter. Which by the way almost never gets discussed here. This is compressor and mic pre central here. Once you have a good relative level on your overall vocal take (this process is obviously nbot exclusive to vocals) I might be tempted to use a touch of compression if needed. Once a track has been properly automated compressing above -2 or -3 simply becomes a matter of tone not any real world level control.

 

[Ford Van]

I think I'd work with volume automation first using a meter and picking a random "average" level. Say -8 to start. From there I try to seat the vocal track (that is staying with the average level but raising or lowering it in it's entirety) ignoring the times it's to much or too little for now. Once you find an average level that sits okay I'd begin the process of automation. Easy enough to automate in segments of whole verse's and chorus's or by whole lines or once you get good at it by individule words.

This of course requires a good meter. Which by the way almost never gets discussed here. This is compressor and mic pre central here. Once you have a good relative level on your overall vocal take (this process is obviously nbot exclusive to vocals) I might be tempted to use a touch of compression if needed. Once a track has been properly automated compressing above -2 or -3 simply becomes a matter of tone not any real world level control.

This is mostly good advice, but who needs the meter? Just volume automate to make it fit. Simple as that.

Compression is more of a "sound" for me in studio recordings. It is not always an appropriate sound either. So, don't depend on a compressor to adjust volumes within the track, unless the compressed character is sought after in addition to needing to regulate levels.

 

[SouthSIDE Glen]

This is mostly good advice, but who needs the meter? Just volume automate to make it fit. Simple as that.

Compression is more of a "sound" for me in studio recordings. It is not always an appropriate sound either. So, don't depend on a compressor to adjust volumes within the track, unless the compressed character is sought after in addition to needing to regulate levels.

Agreed on all counts; I was negligent in forgetting to mention this option.

A nice shorthand way to think about it (very general): Compression is for fixing the vocal, not for mixing the vocal.

G.

 

[masteringhouse]

I agree with Mr. Hanna. For an issue where they are basically two separate parts, try automation first. The compressor settings are going to vary greatly between a screaming vox and a quiet one. Another thing to try is to put them on separate tracks and adjust the volume between the two tracks. Kinda simple solution.

 

[SouthSIDE Glen]

Another thing to try is to put them on separate tracks and adjust the volume between the two tracks. Kinda simple solution.

Tom, I'd like to hang a bit of a qualifier on that one as there are two assumptions implied by that technique: that the vocals want or need to be at a consistant level all the way through the song, and that the arrangement of the rest of the mix is such where the vocals should be at the same level all the way through.

Maybe for the 90% of kids on this board who do all their time hammering out some heavy, radioactive isotope of metal those may not be common concerns, but for those who expand into the myriad of other genres of music that include little things like dynamics and arrangements, those two assumptions can sometimes knock the wind out of the vocals mix.

G.

 

[demoniobox]

im recording pop,classic rock music..like KISS
how can i use automate ?

 

[pod4477]

exactly, compression is for fixing the vox, which definetly helps but not for mixing.

 

[masteringhouse]

Tom, I'd like to hang a bit of a qualifier on that one as there are two assumptions implied by that technique: that the vocals want or need to be at a consistant level all the way through the song, and that the arrangement of the rest of the mix is such where the vocals should be at the same level all the way through.

Maybe for the 90% of kids on this board who do all their time hammering out some heavy, radioactive isotope of metal those may not be common concerns, but for those who expand into the myriad of other genres of music that include little things like dynamics and arrangements, those two assumptions can sometimes knock the wind out of the vocals mix.

G.

Oh I don't disagree and wasn't implying that they should be at the same volume. Only that they need separate treatments in regard to levels and compression. Separating them out this way allows you to do this more easily and makes more logisitic sense when mixing. It also allows you to fix a level issue in a more global way rather than automating a single track all over the place.

 

[SouthSIDE Glen]

Oh I don't disagree and wasn't implying that they should be at the same volume. Only that they need separate treatments in regard to levels and compression. Separating them out this way allows you to do this more easily and makes more logisitic sense when mixing. It also allows you to fix a level issue in a more global way rather than automating a single track all over the place.

OK, I misunderstood you the first time. You're right.

I'm 0-2 today on this board. My brain ain't right today (even more so than usual ). Time to step away from the desk for a while...

G.

 

[Joseph Hanna]

This is mostly good advice, but who needs the meter?

Thanks...!

I'm glad I could deliver "mostly good" advice as the alternative would most assuredly be "mostly bad" advice!??!

To address the "who needs a meter" question I'm really not sure where to begin but two quick scenarios would be...

A) Those who need to, or even wish to, understand and adhere to industry standards.

Perhaps more appropriate on this forum however,

B) Those who wish to better understand the dynamic's of sound as it applies to mixing. Countless times......COUNTLESS TIMES I have been fooled by the energy in a region of audio. My mind and ear tell me the level is low but a quick glance at the Dorrough's tells me a completely different story. Of course by NO means am I advocating mixing by the use of meters. Indeed that will always be the task of trained ears.

As far as automating levels (in this case vocals) I wasn't suggesting setting levels and leaving them. I was proposing setting levels as a starting point and finding a median. A "sitting point" if you will. From there I would encourage setting the level for EACH and EVERY word of the song....if necessary as provided by the composition at hand.

That said absolutely fool-hearty to think your ears are better than a good pair of meters all of the time for all tasks. Perhaps my greatest aid to mixing is my meters and what they provide me in decision making during setting initial levels and ultimately final mixing.

Meters have been an intricate element in mixing since the early days of recording. It's only been recently with the advent of ENDLESS discussions about this mic pre's and that compressor's and converters and tube mic's and Pro Tools-vs-Nuendo and ad naseum has the fine art of understanding, harnessing and ultimately mixing audio been discarded.

Sorry but if you haven't introduced yourself to proper metering there's a hole in your technique.

I'm just saying

 

[tamoore]

I'm constantly needing to 'fix' my vocal tracks. I'm not a great singer, but I do OK on my material.

I compress the hell out of my voice. It really does tighten up the sound (for my voice), and it gets the vocal tracks to lie in the mix better.

I go 10:1, and set the threshold so only the absolute most quiet sections aren't seeing gain reduction. Then, I set the track fader at the level needed for 90% of the track.

I'll also automate in addition to the compression, but only to boost in sections that are needed to be a little louder....

I'm not a naturally dynamic singer, so I need some help. I'm not ashamed to say it.

Is this the 'right' way? Hell - I don't know. I trust my ears. If it sounds good, it is good..... Right?

 

[SouthSIDE Glen]

That said absolutely fool-hearty to think your ears are better than a good pair of meters all of the time for all tasks. Perhaps my greatest aid to mixing is my meters and what they provide me in decision making during setting initial levels and ultimately final mixing.

After having had a good night's sleep and having recovered from the debacle that was my head yesterday (I'm blaming the full moon, BTW ), allow me to provide counterpoint.

Joe, I respect your opinion and really am not out for an argument (honestly), but I feel obliged to take the opposite viewpoint on this one as I heartily disagree with the approach you describe. It may work fine for you; if so, more power to you, that's great. Allow me to explain/describe the alterantive; that for some of us, meters actually hamper the mixing process and hamper our ear's sensitivity to what's really going on.

You state that you're not advocating mixing by sight, but then you go on to describe a procedure that does exactly that; the idea of a "rest level" on the meter is - by definition - mixing by meter rather than by ear.

There are some things and process that have to take visual curs, no doubt. Waveform editing, some threshold-setting techniques, determining and setting headroom, finding specific noise components via FFT, etc. are all methods for which visual cues from visual displays are indespensible. But once gets to the mixing itself, save the visual cues only for that stuff and for keeping us out of clipping land. I believe that using hard numbers or display points as actual "recipies" to follow in the mix does fine to make an oscilliscope happy, but has virtually nothing to do with actual sonic quality of the mix.

I was going to add several other points here about how the idea of a numeric or visual "rest level" straightjackets the mix but I'm going to skip past that and go right to the most important point, IMHO: the more we depend upon sight to do our mixing, the duller our ears become. The duller our ears become, the worse our mixes get.

I'm going out on a limb here, Joe, but I'm going to guess that you have a sgnifigant history in audio enginnering for pro video. In that world the audio needs to be packaged to certain standards much more stringently than it does in the audio-only world; AE/V engineers have been brought up in techniques that are by necessity far more numerically dependant and restrained. AE/V also tends to have a slightly different variation on the definition of "getting the mix right" where the emphasis is more on equal opportunity for the individual tracks than it is on an artictic sonic structure.

Under those conditions, the use of metering as you describe makes much more sense. But I submit that in the world of pure audio engineering, sharp ears and trained brain are the keys to a geat mix and will go much further than any meters can to telling you whether you have the levels correct or not, and that, in fact, extensive use of and trust in one's visual tools over one's ears is self-defeating for the pure audio engineer.

Just one man's IMHO, not a personal attack. YMMV; your method it may work fine for you and others, I'm not saying it doesn't. I'm just saying It'd never work for me, and that there are others who should know why it may not be the best for them either.

G.

 

[Ford Van]

I'm just saying

So, if the singer sings a note with a fundamental frequency of 1khz, then another of 250hz, you would just go ahead and make them the same volume based upon what your glorious meters say?

Careful how you answer.

 

[Joseph Hanna]

So, if the singer sings a note with a fundamental frequency of 1khz, then another of 250hz, you would just go ahead and make them the same volume based upon what your glorious meters say?

Careful how you answer.

Naw man.... indeed you're right. I'll try to be careful.

Then again sticking with the "answer carefully" theme I believe I 've obviously come to the wrong place.

My bad. No need to get up.

I can find my way to the glorious door.