Used for "How to Talk", as stated by mix engineer Kesha Lee in this December 2017 Sound on Sound article.

“The intro track was a voice note Uzi had recorded on his phone, and he played it in the booth on his phone, and we recorded it through the mic like that. It sounded really telephone-y, which we wanted, but I tried to take out some of the highs with the Waves OneKnob. All audio vocal tracks apart from the intro have Antares Auto-Tune as the first plug-in. Uzi hears his Auto-Tuned vocals in his headphones while singing. We used to just have it on default, with a Retune speed of 20, but lately he has been like: ‘Give me more Auto-Tune!’ so now we have the Retune Speed set to anywhere from 12 to 5. The ‘D’ after Auto-Tune is the Waves De-Esser, the ‘Q’ the Avid EQ3 seven-band, and the ‘1’ is the Waves C1 gate. All vocal audio tracks also have sends to the delay and the reverb aux tracks. The delay aux track has the Avid Mod Delay II set to half notes, with feedback at 43 percent, the Waves Renaissance Reverberator, set to ‘Hall 1’ reverb, with the highs cut on the reverb EQ, and the Waves S1 stereo imager. The reverb aux has the Renaissance Reverberator.

“All audio vocal tracks go to the vocal aux track. I had two vocal aux tracks in this session, because I wanted to try something different, using plug-ins I don’t normally use. That’s why one of the aux tracks is muted. The vocal aux track that I did use has the Waves De-Esser acting around 4230Hz, then the EQ3 seven-band which has a high-pass at 96.4Hz, and I’m dipping out muddiness at 200 and 500 Hz. I’m also adding some high end at 6.52kHz. I don’t normally add EQ with the seven-band, but Seth would add some high end on Future’s voice and that worked well, so I tried it here. Next is the Waves Renaissance Compressor, to keep the dynamics in check, and then the Waves SSL E-channel, on which I am again dipping out various frequencies. The latter plug-in is more for colour and character. The Waves CLA-3A is more for the sound, and the RN Digital Detailer made Uzi’s vocals sound fuller and wider.

"The final plug-in in the chain is the Nomad Factory MCL-2269 limiter and compressor, again for the sound and for more volume. We always go for a warm, full, loud, in-your-face vocal sound, also because we like the vocals to be louder than the beat. I always turn the beat down 1-2 dB. There are no plug-ins on the master track, because I used to work for a producer who didn’t want that, as the mix would go to the mastering engineer. So I’m still used to doing it like that. I turn the master volume down anywhere between -7 to -9 dB before it goes to mastering, so they have room to work with.”

Used for the vocals on "Big Girls Don't Cry", as stated by mix engineer JJ Puig in this November 2007 Sound on Sound interview. An image of the settings for the breakdown can be found here.

Vocals: Digidesign delays, Waves Trueverb, Focusrite D2, Neve 1073, Inward Connection TSL3

“The two channels on the far left are the actual vocal tracks [‘LEAD**], and more to the right you can see that there’s a dedicated vocal channel for each section of the song, so I can create the right feeling from section to section. So the first verse [‘1stversvrb’*], the bit with just the acoustic guitar backing, has its own reverb using a Waves Trueverb, which is great. I’ve rolled off above 1k and left all the low end to make it darker and fuller sounding. I’m a Waves guy, they make the best stuff! The D2 gets rid of any dirt below 111Hz and also adds a bit of high end around 12k.

“The breakdown section has a Digirack Mod Delay II slap delay. There are different ambient feels and different attitudes for different sections. Towards the end, where she’s ad libbing, I put a longer delay on. The delays change for each section of the song, but the main long delay that was essential to how the vocals feel is on the entire vocal, which is why I had it on the main channel. It’s again the Digirack Mod II delay, set to 265.49ms.

The rest of the vocal effects were all done outside [the DAW]. I had a Neve 1073 between Pro Tools and the console, because I really wanted the sound of that Neve. The closer you can get to the original source with your effect, the more you’ll get the effect you want. If I’d stuck the Neve on an insert in the Focusrite, the signal would already have been coloured by the console, and the Neve sound would have been less obvious. I probably added +3dB at 12kHz on the Neve and +2dB at 220Hz. After that the vocals went through an Inward Connection TSL3, which is a great vocal compressor that adds great attack. You can get very aggressive with it, and you won’t hear it. I had about 8dB of compression, which is quite a bit, but as I said, you don’t hear it. It made her vocals more expressive.”