Copy link to clipboard
Copied
I am doing the following to edit my audios
What else are the recommendation to make my voice sound even better?
Copy link to clipboard
Copied
Haven't heard the sound of it, so we couldn't even begin to tell you... the only recommendation I'd make is not to use Adaptive NR, because it's nowhere near as good as using the process version with a suitable noise sample.
Copy link to clipboard
Copied
Thank you very much for the response. Below is the link of my YouTube which I have uploaded jsut yesterday and have applied the exact applications mentioned above. It's in Urdy, but still you can hear the sound.
Could you pelase explain further why Adaptive NR is not recommended and what's the option then to reduce noise?
Copy link to clipboard
Copied
Can't see a link in your reply...
The reason that Adaptive NR isn't as good as the process version is simply because it isn't consistent. The effect it has will alter all the time your audio changes, and these changes are generally audible. The main purpose of it is for treating things like film shoot audio, where there's a lot of noise and it alters around a set - it will cope with this well, but it doesn't cope anything like as well with audio where the background is constant and the wanted sound varies.
The process version of NR copes with this much better - especially at the start of a piece where the Adaptive NR is still 'learning'; it works better because it's already learned what the background noise is without having to separate it from the wanted signal - because you've already taught it. So it's this one we're talking about:
I've made a selection, and if you right-click on your noise suggestion, you get an option to capture it as a noise print, which the Process NR will use. (Don't look too carefully at the one in the picture - it's not a particularly good choice!). The other advantage to this method is that in bad cases, you can apply it several times with different FFT sizes, just taking off a little bit each time. This always works better than a single swipe at the noise using just one FFT size.