To denoise your vocals isn’t hard but it’s necessary to even gain the sound levels to even have most possible best source to use to put your effects on.
So get down to business open your projects already placed vocals. Then edit the file pushin slowly on the wave button on the left of each sample:
It will open the Edison: For beginners, it will open just plain Edison like in the first picture:
So you don’t see anything of the noise; Then do as following right click on the wave and select; dual view.
Now you can select the range where the vocals are less visible, also the range without red or yellow spots; Use middle mouse wheel to look deeper into the vocal. Also the lucky horseshoe down on the right side will give you possibility to grid choose or detail choose. Turn it off for detail.
You then gotta choose Aquire noise profile “alt+u” of the red you have covered. Later on mouse wheel out when you have choosen noise profile. The profile does not need to big. Now also choose the whole grid of vocal with mouse button as seen on this; then choose cleanup denoise “Ctrl+u”
Now the trick part; choose rather threshold to bottom; that will ensure lowest sound level will be affected by noise removal. Also the denoise level of your denoising. You can set it to light 10-30. Depending on how you want to do it. Sometimes double up denoise is required. Depending on the quality compared to noise profile you have. Some might tend to use a harder level when the noise is just noise. But when noise is instrument and the vocal is vocal. A very lighter removal is needed. Then at least 10+ times removal. Mostly because when denoise levels are to hard there will be some artificial sound in the high frequencies.
You can also play the track, but you have to remember your levels if you are doing precise denoise. Since you might end up denoising a whole track for example filled with instruments, that way your vocals start at the middle of the track. It’s possible but for enthusiasts using all theyr whole day trying to get best possible Madonna vocals.
Also try using the Output noise only to hear what you are removing, are there vocal notice you probably have a wrong noise profile.
Also the frequencies from 100-1000Hz can often be easy removal and hard on your vocals bass feel and might cause different aspects of removal. Causing different problems. Since the waves are not that easy to recognise just hearing the vocals in low depth frequencies.
Remember to always save your original vocals first when your record with Edison!
As you see here, there is just slight removal with threshold -6dB and denoise level 22.7. So I was wondering to just turn up the next denoise level to around 30 and throw in the last denoising for this vocal. And I remembered to aquire new profile; but somehow the noise was very artificial since I actually missed with the original and the old denoised vocal from 2008. So I did throw a level of threshold up to around 0dB. That way most of these white dots disappeared in the second picture; But in your track you have to focus on the noise within your vocals. Different levels of threshold does not need higher levels since noise most often are at ranges of -60dB already.
Did you know that noise on anything out there is the problem of 24bit and 16bit processing that Windows computer has? Especially when FL Studio is 32bit main processing. So many apparats do create problems for you anyhow. Its real pitty tho. Hard workship make a pitty sound. But remember there is an solution to that; And that’s the newest Creative AE-5 sound card. With 100% 32bit processing.
Since when you don’t dither in FL from 32bit to 16bit you also get noise. That is the particualar noise that is created. So there is perhaps no soundcard yet as only Creative for FL Studio. Always in the end remember to use dithering on anything being 16bit. 24bit in FL Studio is also an lost case.
So when I want to save the file; It will be saved in the sliced beats folder in; When I drag and drop the vocal that is marked red in Edison on playlist or step sequencer.
C:\Program Files (x86)\Image-Line\FL Studio 12\Data\Patches\Sliced beats
You might want to remove the artificial sound on an already processed by others vocals. But these also most often occur at higher frequencies above 5kHz. But in this that I worked with my vocals weren’t sounding loud than 5kHz. So I chopped it down.
It’s possible to gain further new frequencies of higher than 5kHz. If you lost it all. With simple exite technique. Since making soprano sound is a possibility to do so. If you need that kind of knowledge it does not come for free. It’s beyond what most would imagine possible. But for me it’s like doing the lossless process on an mp3. It’s different excitement processing for the high frequencies. It’s also God like process, since much of the process is out of logic.