Hey gang,
In response to a few emails I got on how I recorded and mixed vocals
for a song I posted on this forum the other day, I thought I would share
the play by play for vocal tracking and mixing and my methodology
for you to check out, comment on, tear apart, rip me a new one, whatever.
The song in question was mixed using the things I learned from Charles Dye's
HDL series in Digizine. If you'd like, you can listen to the song by clicking this:
www.halftheworld.cc/time.mp3
I'll give you the full run down on the exact step by step playback on
those vocals and the mix in general, as well as a very cool trick for
printing your mix and burning a cd without ever using Pro Tools bounce to disk
(which many people feel degrades your sound quality).
First, Mary sings the song from top to bottom on 6 separate vocal takes,
into a Studio Projects C1 mic with the switch in the flat position.
The mic is connected via a Monster Cable Performer 500 series
21 ft. mic cord into the Apogee Trak 2. The Trak 2 does not use Soft Limit,
nor any ASP functions. The vocal is sent at 24 bits straight into PTLE
with the levels peaking into the yellow, but never into the red.
Now I have several vocal takes which need to be edited. On the song
"Time" as well as all the Half The World songs I'm working on, I edit
vocals the old fashioned way . . . and it's very time consuming . . .
it takes hours. I take the first phrase in the song, and listen to that
phrase and audition each vocal take, listening only to that phrase,
against the band playing and decide which vocal take offers the best
performance, and I cut and paste that phrase into a "comp" vocal track
using playlists. Sometimes I cut and paste individual words.
Hours later, I have a killer sounding composite lead vocal track.
Now I start over on a new track which will be the "comp" double track vocal.
Until I discover a faster / better way to do this, I'll be stuck using the old fashioned
method. Once I have a comp lead vocal, and a comp double track lead that
matches it as closely as possible, I'm ready to mix.
When I mixed "Time" I used a standard set of plugins which have always given
me a huge present vocal sound and I set them up the same way on every mix
even for other bands singers . . . as follows:
1. In the first insert position, Auto Tune, set to the key of the song, in automatic mode,
with appropriate notes dialed in or out accordingly, so that there are no weird moments.
2. In the second insert position, Waves Renaissance Compressor set to the Vocal preset,
then I lower the threshold so that the loudest moments of the vocal performance sound
smooth, but not squashed . . . this tends to run anywhere from 3 to 6 db of gain reduction.
3. My secret weapon! . . . Waves Renaissance Vox, this is where the magic happens. I always
use the 25dbcomp preset . . . I can't explain what this does but it makes the vocal track much louder
and fuller, much more present and intimate . . . then I obviously have to back the volume of the vocal fader
down a bit to compensate for the greatly increased gain. This is the best preset ever ! It creates
a vocal pocket almost on it's own!
4. That's it for plugins on the lead vocal. Now, on the lead vocal track I create a send to bus 1-2 (or whichever)
to the Reverb channel which is a stereo aux input with an appropriate verb dialed in. I adjust the amount
of the send to taste.
5. I duplicate Auto Tune and the Ren Compressor plugins with identical settings on the double track vocal channel,
but not Ren Vox . . . this goes on the lead only.
6. I now adjust the fader level for the double track vocal channel so that it's about 60% of the lead . . . in other
words I try to make it sound like one full voice rather than two voices. Also, I may mute it during the verses
then bring it in during the choruses to make the hook stand out . . . just depends on whether it works on the
verses or not.
7. If I want delay or effects, I'll usually NOT put it on the lead voc, but rather do the old trick of putting a send on the
double track vocal, which routes over to an echo or flange channel, with the effect dialed in at 100% wet. This gives
a nice effect on the vocals but it's not effecting the lead, which makes the lyrics remain clear and audible.
That's it for vocals!
Now for a very cool trick you can do with your Apogee (or any device with digital ins and outs) . . . you may already know about this but . . .
When it's time to print your mix, don't bounce to disk, instead send the whole mix through optical 7-8
out to the Apogee Trak 2 with UV22HR turned OFF (no dithering) .
Simultaneously, have a new stereo track armed and recording and MUTED with it's inputs set to Optical 7-8.
Name this new track 24mix (because it's an undithered 24 bit mix).
When the song's done, you now have a 24 bit stereo mix on a stereo track.
Save and Close.
Now create a new 24 bit session. Import your 24mix stereo track into this new session. Create a new stereo
track named 16mix, set it's input to Optical 7-8, MUTE IT, and arm it to record. On your 24 mix channel, insert
any plugins you would normally have put as a last step on your master fader, light compresion, etc. Create a
Master Fader and send your 24mix track out to the Apogee with UV22HR dithering turned ON for 16 bits (or a dithering plugin).
When the song is done, you have recorded a 16 bit dithered final mix. Don't don't touch the 16mix channel or
you'll add wordlength.
Save and close.
Open up Jam and drag the two 16mix files straight into the Jam window and it will show up as a single song.
Burn your cd!
This is one of the best ways I've found (short of a pro mastering lab) to get burned cd of your mix which is virtually
identical to the original 24 bit files . . . bypassing the Pro Tools Bounce to Disk feature.
Sorry for the length, I thought it might be helpful and fun to discuss.
Pete