Thread by @alyssilya

_PART 2! Quick guide on making my pro vocal chain entirely for FREE! Treat your voice right and learn to master your singing~ 🧵 #vtuber #ENVtuber

Quick disclaimer: while this setup was made for singing both in the studio and live, I don't recommend building this directly in OBS if you want to use this on stream (its not impossible just a pain). I use an external DAW, and in part 3 we'll look at how to set one up for OBS.

(disclaimer cont.) If you want to go with OBS anyway, make a new mic channel and use the method from this thread to add the plugins:

https://twitter.com/alyssilya/status/1531417197700124677?s=20&t=ddsr66RfMOshKjkDSpjHPA

This will give you one "clean" channel for speaking and one with this setup for singing.

Let's get started. Quick refresher on the signal chain we're replicating. READ PART 1 FIRST IF YOU HAVEN'T, that's where I broke down what everything here does:

https://twitter.com/alyssilya/status/1536888289084465154?s=20&t=ddsr66RfMOshKjkDSpjHPA

We'll be using the kHz plugins from the voiceover guide to do the bulk of the work, more info here:

https://twitter.com/alyssilya/status/1531417213210615814?s=20&t=ddsr66RfMOshKjkDSpjHPA

First thing to tackle: pitch correction.
Most solutions for autotune are pricey and resource intensive, so introducing Gsnap by GVST! Completely free, runs pretty light, and the pitch tracking is surprisingly accurate. Just set they key you're singing in! gvst.co.uk/gsnap.htm

Next we'll want to build our deesser. Biggest change from the voiceover guide is turning the attack down on the kHz Transient Shaper, but beyond that it will be pretty much identical. Follow this part from the voiceover guide:

https://twitter.com/alyssilya/status/1531417240230313984?s=20&t=ddsr66RfMOshKjkDSpjHPA

Next up, add a Khz 3-Band EQ. Cut the lows pretty much entirely, move the mids to surround most of your voice, and boost them slightly. After that, literally just copy-paste the compressor from the deesser stack and lower the ratio.

Next up, we'll be using Xfer's recreation of Ableton's OTT. It's functionally identical to the original, we're just gonna cut the lows almost entirely.
Available for free on Splice: splice.com/plugins/3788-o…

Right, so on the original chain there's 2 comps running in parallel. As mentioned in part 1, I use 2 because Ableton comps even at max ratio don't have a lot of force and I need 2 to do the job of 1. To avoid complex routing, we're just gonna use one kHz Comp. High ratio, +makeup

These next 3 components are optional, but they add a lot to a vocal. First up is kHz Distortion. We'll be using the Overdrive setting. Lower the Dynamics to around 25%, push the Drive to +6 dB, and add some Spread.

Next plugin is a kHz Convolver. Load up an IR you think sounds good but isn't overpowering and doesn't wash out the sound, just adds a bit of space. Turn the mix down if needed. Add a kHz Reverb if you want to add a more washed out sound.

Follow your reverbs up with a kHs Stereo. This will make the mono vocal signal into stereo and give your voice more width.

Last step! Normally I top everything off with iZotope Ozone on my VOX bus before it hits my master chain, and the main thing I use from it is the limiter. For our free build we'll use the Khz Limiter. Sing loudly, lower the bar until you're out of the red on your mixer.

That's everything! Our final signal chain should be:
- GVST Gsnap
- Deesser stack
- kHz 3-Band EQ
- kHz Compressor
- Xfer OTT
- kHz Compressor
- kHz Distortion
- kHz Convolver
- kHz Reverb (optional)
- kHz Stereo
- kHz Limiter

Remember to test your mic latency & offset in OBS!

Part 3 will cover making a DAW play nice with OBS and untangling all that fun audio driver bs. Drop a follow so you don't miss it, and thanks for your support recently! It really means a lot to me 💙