I was hoping for some more comments from J. He is probably too busy so here comes an explanation of what I did.
First up, te recording and performance are very nice. This, in a sense, was more challenging than having loads of tracks to play with. Things had to be right because there was no where to hide.
I listened to the material to get an idea of what to do. I don't listen to much music with vocals so I asked my girlfriend for something. She gave me Christina Aguilera - Hurt. That helped me decide to go for a big pollished pop sound.
Techniques: Tracks loaded into Sonar 6. Piano panned hard left and right and sent to a bus. Room track not used mainly because it was mono and it would be harder to integrate it into an artificial reverb instead of just faking the whole space.
Vocal pre-processing:
First I loaded the vocal into SoundForge and hipassed all the pops at arround 300Hz. I also selected all the big ticks and used the smooth function in SF to remove them.
Vocal chain:
- PSP MasterQ Hipass at 120 Hz 24 dB/Oct. HiShelf at 10Khz + 2.8 dB.
- Voxengo Voxformer in dual band mode. Cross Over at 1.4Khz. About 2 - 3 dB of compression on the low band. 0.5 to 1.5 dB on the high band. De-Esser set to 6Khz about 2 dB of De-Essing. Alot of the "Presence" knob but set to 100Hz. This give back the weight that was hipassed by the MasterQ.
- VC-64 Vintage Channel 3 - 4 dB of De-Essing at 3.3 Khz. 2 - 4 dB of compression quite fast attack, slow release, ratio 1.4:1, smooth mode. Another HiPass at 91 Hz, 3 dB boost at 160Hz 3.4 dB Boost at 5Khz (Q 0.7 in both cases).
- Sent to Global Reverb and Vocal Reverb.
- Output to Master Bus.
I should have done some volume automation so that it started off a bit softer and grew with the song. Next time better.
Vocal reverb
- Wizooverb W2 set to Choir Cathedral IR with 4.2s time.
- PSP MasterQ: Hipass at 60Hz, LoShelf at 840Hz -2 dB, -1.5 dB at 250Hz Q 0.4. LoPass at 7.8Khz 12/Oct.
The vocal reverb goes to the master bus but also to the global reverb.
One thing I should have done is set the vocal send to the global reverb pre inserts so that the reverb gets bigger when the vocals gets bigger. I forgot ... :/
Piano bus:
- VC-64 Vintage Channel: Between 1 to 3 dB of compression (Attack 14.5ms, release 100ms, ratio 1.5:1, mode smooth and optical. Hipass at 30Hz.
- PSP MasterQ: Another Hipass at 30Hz 24 dB/Oct. LoShelf at 700Hz -0.6 dB, -2 dB at 270Hz Q 0.4, +2 dB at 90Hz Q 0.3, +0.7dB at 6Khz Q 0.3
- Output to master bus and send to Global Reverb
Global reverb:
- Wizooverb W2 set to Large Chamber IR with 1.6s time.
- PSP MasterQ: Same as Vocal verb EQ.
There is a short amount of delay at one point on the vox. I just used the Sonitus delay for that.
Stereo bus:
- Waves SSL Stereo bus comp: Between 0.5 - 3.5 dB of compression. (Attack 1ms, Release Auto, Ratio 4:1)
- Voxengo Warmifier set to 6550 emulation. (For fake analogue warmth)
- Voxengo Warmifier set to 12AT7 emulation. (For more fake analogue warmth)
- Voxengo Elephant set to EL-3 mode. Max 0.7 dB limiting on the left channel at the loudest peak. DC filter 18 Hz Butterworth which gives a nice sound to the low-end IMO.
r8brain free for sample rate conversion to 44.1Khz.
Alistair