A breakdown on which plug-ins I used to mix the vocals on our latest single 'Chosen' feat Jessica Martins. Check out the track and our mixing methods below:
Jessica has beautiful soft tonal qualities, I wanted all the intimacies and warmth of her vocal to shine through in the mix.
1) I generally start my processing chain with some mild compression - in this case the CLA-2A, my usual favoured choice. It's modelled on an old favourite among engineers, and immediately makes the tone of the vocal warm and inviting. I usually set the attenuation to around -3.
2) The compression from the CLA-2A tends to be a bit slow in taking action, leaving some transients untouched, so next I inserted the Waves DeEsser to control the sibilance - I inserted this before the CLA-2A. DeEssing at this point is nominal, but becomes critical further down the processing chain as we insert more compression. Remember, dynamics can play a huge part in a vocal performance, sometimes as great as 30 to 40 DB difference, yet the transients don't really change in level. In other words, whether a vocalist shouts or whispers has an obvious huge change in level, yet the difference in transient levels are nominal, if none. You can't shout an 's' or 'f'' or 't' :) So in essence, I try to treat the sibilant letters as their own entity as early in the mix as possible. In this case I set the attenuation to around -6 and used the split mode. You can play around between split and wide mode to see which sounds better. Split mode kicks in above the set frequency, whereas the wide mode keeps the frequencies, but brings down the level of transients.
3) Next I used my favourite tool for vocals, the Waves Renaissance Channel. In a nutshell, the RChannel is a complete channel with a gate, compressor and EQ - you can change the order of compression and EQ. There is good debate surrounding this topic, traditionally EQ is applied first. Personally, I get best results with applying compression first, then EQ'ing. At this stage in the processing chain, I tend to compress the vocal a lot, ironing it out as much as possible. I then use the EQ to filter, 'colour' and shape the vocal evenly. With Jessica's vocal I selected the RVox (it has a softer knee than the Rcomp) and lowered the threshold a lot, setting the attenuation to around -10. This obviously makes the vocal softer, so I adjusted the makeup gain accordingly.
4) I next used the EQ of the RChannel to colour the vocal. Setting the curve and identifying frequencies is the most sensitive and time consuming part of my mixing process. In this case I cut unnecessary frequencies below 180Hz, boosted the mid range at 3.5KHz for presence and 'shine', with a relatively narrow Q. At this stage the vocal sounded a bit thin, so I then boosted low mids at around 300Hz. I also brightened it up by boosting around 9KHz.
5) I next inserted the Waves Vocal Rider, another of my favourite plugins. I really enjoy using the rider - it brings out the endings of the words and makes the vocal so much more intimate. Once I was happy with the result, I let the Rider write the automation curve into my sequencer, and then tweaked manually wherever necessary.
6) Next in my processing chain was the C6, a multi-band compressor. I used this to gently compress frequencies that are 'out of control'. For male vocalists this tends to be around 250Hz, and for females around 500Hz. Compressing these frequencies will help the vocal 'sit' in the mix without compromising the strength and power of the mid range. I usually curve out a relatively tight notch using the Q, set a fast attack - then play around with the range and threshold. In this mix I also used the C6 to further de-ess by compressing around the 8KHz mark.
7) Time for effects. I started by sending the vocal to an auxiliary and inserting the Waves Rverb. I typically don't want any early reflections as these would place the vocal in a small space and make it sound 'roomy'. Each mix is different, but with Jessica's vocal I switched the early reflections off altogether and just concentrated on the tail end of the reverb. The pre-delay plays an important role; no pre-delay will place the vocal further back, a longer pre-delay will bring the vocal forward and make it more present. In this instance I set a relatively short pre-delay of around 9ms, and set the tail to around 1.4 secs.
8) Next I used a stereo harmoniser, the Waves Doubler2 - inspired by the AMS. I EQ'ed the send to boost the top frequencies, I did this by inserting an EQ just before the Doubler and boosting a tight notch around 9KHz. There are more options here - you could use an aural exciter to dirty it up a bit. The parameters on the Doubler are amazing, lots of modulation to play around with. In this mix I kept it simple, slight detune of +6 and -6, and a short delay of 9ms and 23ms on the right and left. It's important to note that this should be used as a very subtle effect, less is more. I use this effect to give the vocal just a bit of sparkle.
9) Next up I applied short delay using the Waves SuperTap. I panned to either side and filtered out the lower frequencies, everything below .6Khz. Once again, I used this effect moderately. The repeats tend to be a bit dry, so I sent them to the reverb auxiliary to blend in a bit more. I also inserted another de-esser before the Supertap to control the sibilance.
There you have the basic elements of my mix. I used the above chain for the main part of Jessica's vocal, however the chorus needed to sound wide and required some added effects which I haven't mentioned. Please feel free to comment and let me know what you think - I hope it's not too obscure and that it makes sense.