vocal mastering problem

  • Thread starter Thread starter MTY_MEX
  • Start date Start date
M

MTY_MEX

New member
It seems that I have a consistent problem mixing/mastering vocals over karaoke tracks... I thought I'd finally got a recipe to record vocals and polish them before mixing to karaoke tracks.. but I feel frustrated again and looking for help..

My setup and chain:

To record: m-audio condenser mic >> small COMP16 compressor >> small mixer UB802 >> laptop
To play karaoke and listen when recording: laptop >> UB802 mixer >> headphones

process:

once having mixdown of one single vocal track >> EQ (cut -18db below 80 Hz, boost 3db at 13,500 Hz Q=0.8, cut 3 dB at 180 Hz, Q=90.5, cut 3 dB at 1800 Hz, Q=0.5) >> vocal compression (sometimes i use the "Vocal Compression" preset of Izotope Ozone, sometimes waves Rvox then waves C1 with almost identical results) >> waves Rverb (I don't know which preset to use: vocal plate, studio A and which wet/dry settings, but I normally use vocal plate at 60 ms predelay, 2s reverb time, 20 wet/dry) >> Normalize to 100% >> Apply Izotope Ozone's "Intelligent Maximizer" preset or Waves Ultramaximizer set at -5 dB (with almost same results) then cut this wave -10 dB and mix with the already cut equally -10 dB karaoke track.. then the new mixdown is normalized to 100% then Ultramaximizer "Hi-CD resolution" preset.

I get loudness and some brightness but some parts of the mix sound too bad (the vocal part) and I feel i'm still far far away to what could be called "commercial cd" level

please give some indications / suggestions, I'd appreciate them.

Thanks in advance.
 
That post makes my head hurt. Here are some thoughts:

1) If you are really running through the compressor before you hit the preamp, that's wrong and how do you get phantom power to your mic through the comp? The compressor (if used at all) should be plugged into the "insert" point on the mixer channel.

2) There is absolutey no formula for EQ that works every time. Each song/voice combintion will require something different. Arbitrarily doing anything is a bad idea. Listen to the song and learn if anything needs to be corrected with EQ.

3) You cannot reasonably expect "commercial cd" quality from that setup. If you work hard, you can probably get "usable demo" quality.

4) You might have to EQ the backing track to "fit" the vocal in there better. Again, there are no absolute rules on how to do this. You have to listen to each mix separately and decide what it needs.
 
I'm definitely no expert on this but I agree that you need to go through each track individually and find out what eq you need to do for the vocals to fit in with the music. There is only so much "space" that you have to work with. Also, be careful with messing with the dynamics too much. Compression is useful at times but everytime you compress you're destroying the dynamic range of the song. You need to control the limiter so it gets your music to the level you want but doesn't squash everything. If you do too much it will sound loud, but will also sound like garbage since everything will be pushed to the same volume. It's easy to make anything loud but make sure you're not wrecking the sound at the same time.
 
:D yO Amigo Mex:

Don't know what your "set-up" is; however, you might try this:

Run your vocal mic to a good pre-amp.From the pre-amp, run a line to the compressor and experiment. Limit your reverb. You need reverb on a vocal but you do not need echoing canyons, except for special effects or affects.

You can put your vocal on two tracks and that will give you more boost.

Depending on your gear, you can also boost the stereo tracks with some EQ or stereo input volume. You need to play with it a few times to get what you want.

I've always been a "voice up front" engineer and it took a while to figure out how I could get bigger volume/without clipping and yet have enough music in the background. I don't have one of those multi-bucks devices to do booming volume as found in many CDs; however, I'm pretty happy with the results I get and so are my vocalists.



Happy Spring
Green Hornet :cool:
 
First thing, Mixing and Mastering are 2 different steps. You do the mix first, then mixdown. After you have a final mixed down version, then you master. The goal of the mix isnt getting your final levels loud. Mixings goal is to make the song sound good.

In mastering, that is when you compress the entire song, fix tonal issues, and get the overall level up to cd standard. Also, the mixer and mastering engineer should be different. A mastering engineer has years of experience with EQ and Compression.

With that being said, as for EQing the vocal track, remember, EQ is to fix tonal issues. If it sounds good without EQ, then dont use it. With really good preamps and mics you may not need to eq if the track doesnt tonally conflict with other tracks. If you do use EQ, NEVER EVER add or take out no more then 4dB (unless you are experienced with EQing or going for some sort of effect).

As for compresses your vocal track, use one compressor. If you want a more flat sounding vocal (not as much dynamic range), use a higher ratio (8:1 and up). Again, there isnt any real rules, just what you want. If you are just starting out, use 4:1 compression ratio as a starting point. Quick attack and a normal release.

DON'T COMPRESS YOUR FINAL MIX!!!!!!! Let that happen in mastering. That is the biggest problem that mastering engineers have with new mixers. It really hurts the final mix. Let the pro at compression handle compressing the final mix.

If you get a really good mix, then spend the bucks to have a mastering engineer master the track. It's worth it.

I hope that helps
Matt
www.smackmastering.com
 
Well I suppose I'll comment on this.

First off I'm not a karaoke expert, but I feel it's safe to assume that you are dealing with already mastered music that you are trying to fit your vocals too. So your main concern would be the treatment of the vocals.

The only thing I can see doing to the Karaoke track is perhaps making a slight dip ot to in the eq in places you want to highlight your voice. Those would be places you may want to make a eq spike in the vocal track. Now I'm not talking anything big. pretty narrow bands with maybe 3db + or - gain adjustments. You should only be lowering the whole karaoke track enough to fit the vocals in. ducking both tracks to maximise the mix later is only damaging not helpfull. you should be able to fit vocals over the karaoke track with very little gain reduction to it. If not the vocals are some how too big.
-10db is way too much. at least jack them to the max you can while mixing before any maximiser.

Secondly if you have good copmpressors "in the box" which it sounds like you do, don't wory about compressing your vocals going in if you can help it. It's just that much less fexability you have later and might be bringing up things you're having counteract later.

Don't get caught up compressing & eqing your vocals when they are solo'd. You can use that to get you to a starting point or as a referrance once in a while, but do your work in front of the music. If you need to bring the vocals up a little louder than they will be in the final mix to work on them and then sink them back in later and refine at that point.
My point is don't get hung up on the fact the vocals may not sound fantastic to you solo'd. Who cares.

Last point, I'm for hire & pretty damn cheap :D Shoot me raw tracks and I'll take a listen if you want ;)

You'll get it. You may want to post an example. Better advice would come in the mp3 mixing clinic with that. Check some referrance material to make sure you not striving for unrealistic results ;)

F.S.
 
Last edited:
:D Yo SmackMan:

Why compress the entire song if it isn't necessary to compress the "entire" song.

I'm using the yam 2816 and I think we are saying the same thing using different words.

I record, say 4 different tracks; these tracks I put into stereo tracks in order to make the CD. It is the stereo track where pushes and shoves are made, is it not? As the stereo track, when acceptable to the dial twiddler, me, is what will have "loudness" where it is desired by the talent, writer, or engineer.

I've just messed with a poorly recorded track on a "major" CD by a major artist. I felt the person who did the mastering put the vocalist in a digital clip. So, by moving some faders and blinking the eq just a tad, I got the clip to sound normal. I guess I should send "her" a copy but "they" may not like me messing with their CD. Ahh, that's life.

Hoping spring has sprung wherever you are.

Green Hornet
 
ok, some updates, lot better but still having doubts

Thanks for your kind responses.

I reviewed the software processing and realized that i was doing things wrong (hardware is ok, i missed to mention that mic goes to mixer then mixer fx send output to compressor -which only has hard presets, cannot modify parameters- and then to my laptop).

i was lowering both tracks (karaoke and vocal) -10 db to avoid clipping when mixing down, but i was doing all the job in the EDIT TRACK environment (i'll say it againg for you not to laugh that much: i don't live from this, i'm just a hobbyist that can practice only after my beloved wife and kids go to sleep..) and was applying all what i mentioned step by step affecting the raw track and was moving to the Multitrack environment just to listen the results...

after your posts i did things differently:

1) load and clean the raw vocal track (noise reduction, inevitable as I'm using a mixer instead of a USB interface)
2) load the karaoke track
3) insert both in the Multitrack environment
4) get back to the edit environment to work on the solo vocal track
4.1) apply waves rvox (-3.5 dB)
4.2) apply waves c1 (-7 dB, 1.8:1 ratio)

at this point I get a more balanced vocal track

5) get back to multitrack environment and START USING THE FX BUTTON IN THE VOCAL TRACK
5.1) insert (or send, i don't know the appropriate word) AA delay (10 ms) to thick the vocal
5.2) insert waves Rverb, vocal preset, 15 wet/dry ratio and 1.8 ms for the reverb time
5.2) insert waves de-esser (-18 dB)
5.2 tweak the volume of both tracks until they sound well, the music gets down to around -3 db and the vocal gets up to around 4-5 dB
6) mixdown (which takes me back to the edit track environment)
7) apply ultramaximizer to the mixdown (between 4 - 6 dB of gain)

the result is much better and i don't need to apply EQ

please review this new process and let me know if there is something else to improve.. the overall idea (i also forget to mention) is to make some money recording people that like to sing with karaoke tracks, just for fun, but i have seen that many people that likes karaoke are also willing to have a recording, so I made a small wooden booth (acustically treated) and spent months trying to find out a "standard process" to quickly get good results..

one day i will install and run a formal studio, it is a goal for when i'm retired, (i work in the telecom business) but for now i just want to learn as much as possible


again, I appreciate your advices.
 
You don't need to apply noise reduction on the raw vocal track unless it is particularly noisy. It really shouldn't be though. If you're running the correct gains and levels you should have any noise coming through.

I run the outputs of my mixer into the line in of my card and assuming I set the levels correctly theres no noise at all that you can really hear.

Learn your gear, some settings on my pots are inherently noisier than just a few degrees to the left.

Noise removal plugins fuck with the sound and unless you need to do it there's absolutely no reason to. Same with the de-esser. Unless your getting bad sibilance don't do it.
 
Stop using presets.

Ozone can sound great...but never by using a preset...ever.

Every sound needs it's own settings...use your ears.
 
You must log in or register to reply here.
Back
Top