Input latency minimum is 30ms. How to get lower latency (below 10ms)

Hi,

So, I'm an idiot.

Not once did I think about all the effects I'm adding for the track, even though after finding the below video, I remember reading about the exact same thing.

The WORST offender is Speech Volume Leveler.
An example can be seen here: 

https://youtu.be/IvXoB_jyJVw?t=351

So, yeah.

Now I'm left wondering how do radio hosts sound so good if they can't add effect in real time.

Do all of them have analog gear (equilizer, compressor etc.)?

So, the Scarlett Solo does have an ASIO controller which I have now installed and using in Audition. My headphones are now connected to the Scarlet Solo (instead of the PC) and the PC out is going to the solo.

BUT, the latency hasn't changed.

I am guessing that the fact that I'm routing the PC to the Solo is a No-No, defeating the purpose of the whole ASIO advantage, since Windows is the one doing the routing.

Ultimately latency is determined by the hardware in your computer, and you can't eliminate it. It's caused by the finite length of time that it takes to convert your analog speech to digital, process it through the operating system, and then output it again. In order for the signal not to break up during this process, Audition uses a buffer which you can set to be slightly greater than the system latency value, but that's it - you can go no lower than that - and even with a top-notch system you won't get this below quite a few milliseconds. On your machine, it's going to be a lot worse because you're using an MME driver, but you don't have a choice about this; Focusrite don't appear to have an ASIO driver for the Scarlett Solo. If you had a sound device with an ASIO driver, then this would bypass most of the operating system, which is probably the biggest cause of additional latency there is in your system. So I'm afraid that the latency value is inevitably going to be higher than you'd like, but unfortunately there's no fix for this - it's down to the Laws of Physics.