To denoise your vocals it isn’t hard but it’s necessary to even gain the sound levels, too even have the best possible source to use to put your effects on. Some effects will not have a good quality based on the original recorded source.
So get down to business, open your projects with already placed vocals in its project. Then edit the file pushing slowly on the wave button on the left of each sample or you already perhaps have it in your Edison:
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; Spectrum.
Now you can select the range where the vocals are less visible or with just noise, that’s the range without red or yellow spots; Use the middle mouse wheel to look deeper into the vocal. Also, the lucky horseshoe down on the right side of Edison will give you a possibility to grid choose and easily detail choose. Turn it off for detail.
You then gotta choose Aquire noise profile “alt+u” of the red you have covered with the first mouse button as perhaps the whole vocal sample. Later on, use the mouse wheel out when you have chosen a noise profile. The profile does not need to big at all, even near a few milliseconds just that you believe it’s enough noise to cover the noise spectrum. After then also choose the whole grid of vocal with mouse button as seen on this; then choose cleanup denoise “Ctrl+u”
Tip: If your vocal has precious bass frequencies but as well you see the color yellow or red on your vocals you are trying to use a noise profile. Then all you should do is to reduce this level with something called Equalizer. It is a specific choice especially when you are thinking of to remove certain frequencies that already have a vocal touch or sound that you don’t want to be in the noise profile.
Now the tricky part as shown in the next picture; Choose rather a threshold to the center but the higher the threshold level it is the denoising will be at a higher rate. This better than the denoising level, it does not do that much to all levels. Sometimes even having it at a higher range will eventually give you certain levels of removal. But -6dB is mostly often sound lower than -6dB of noise so setting it to negative value would for some be better, but having it centered is a choice. Most often any noise is lower than even -32dB. But it can ensure the lowest sound level will be affected by noise removal so a higher threshold will ensure a higher probability of removal of detected levels.
Also the denoise level of your denoising: You can then set it to light 10. In the blue color. Then repeat the denoising process 5-6 times. This is near the same with the newest FL Studio. It only works most often for 32bit FL Studio.
But when noise is an instrument and the vocal is vocal. A very lighter removal is needed. Then at least 10+ times removal. Mostly because when denoise levels are too hard, there will be some artificial sound in the high frequencies, that can really make a vocal out of proportions. Be aware! It can turn your vocals sounding robotic!
Remember you can reduce the denoising as shown in the red square, that way you can reduce the denoising done to bassy levels of the vocal. Sometimes there might not be noise there but just particles that aren’t noisy.
Tip: You can also play the track using “Output noise only”, but you have to remember your levels if you are doing precise denoise by going back to the whole track process. Since you might end up denoising a whole track for example filled with instruments, and that is needed to outsource the vocal. Because you can only monitor how the vocal will start to begin with being made denoised. It’s possible but for enthusiasts using all, they’re whole day trying to get the best possible Madonna vocals for that.
Important Tip: Don’t do the denoise on already recorded vocal you have done in Edison. First, make a copy of the track and paste it to your playlist. Using the pointer button and drag and drop that vocal track to the playlist. The recording will then stay in the sliced beats folder. In FL Studio 20 it’s in the Documents folder.
Also, try using the Output noise only, to hear what you are removing. if there are vocal noises you are probably having a wrong noise profile. Be careful with wrong profiles!
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 recognize just hearing the vocals in low depth frequencies.
Remember to always save your original vocals first when your record with Edison!
Did you know that noise on anything out there is also the problem of 24bit and 16bit processing that Windows computer has on recordings? Especially when FL Studio is 32bit main processing. So many units do create problems for you anyhow. It’s really a pity tho. Hard workship makes a pitty sound. But remember there is a solution to that, And that’s the newest Creative AE-5 sound card. With 100% 32bit processing. Especially the best solution for a 32bit FL Sound. Giving the advantage of anything.
Since when you don’t dither in FL from 32bit to 16bit you also get noise. That is the particular noise that is created with different units with 24bit to 16bit change, or 32bit to 16bit. 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 and 24bit in FL Studio- Never do a release by 24bit on FL lower than FL 12 or even 20 without dithering! That is dead noise and destruction. Since FL 24bit dithering does not work properly for all FL Studio before FL 12. You should really upgrade FL Studio. But as well to always dither if you are unsure what dithering really is.
So when you want to save a copy of your file; It will be saved in the sliced beats folder in; When you drag and drop the vocal that is marked red in Edison on a playlist or step sequencer.
C:\Program Files (x86)\Image-Line\FL Studio 12\Data\Patches\Sliced beats
On FL 20 that is in the documents folder.
Tip on denoised robotics: 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. So chopping it down can be a solution by all frequencies higher than 5kHz removed with Parametric EQ 2 preset 20hz cut + 18kHz cut. You can drag to 5kHz. This can remove too much in vocals. But are a good choice for too many artificial sounds.
It’s possible to gain further new frequencies of higher than 5kHz. If you lost it all. With a simple excitement 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 CS12M by Terry West already has a good extra effect for Exciting high frequencies. Mainly copying the sound of lower frequencies to higher ones as masking. Even Modern Antress Plugins can do that magic. If you don’t literally think by logic. Some unlogic changes can end up in magic. It depends on how you know your soulful music. Like makeup of the frequencies can be done.
End Tip: Remember to adjust the denoise amount to max 10. Then use multiple times denoise. Up to 5-8 times. And you don’t need to acquire a new noise profile each time!!!
Just denoise without adding up to artificial problems by wrong settings!: