As above…I have spent a fair bit of time trialling all that I could see in the manual etc and it sounds like it should be easy, but no matter what I do I can hear the effects on the recorded elements
I have attached a test with some of the classical guitar…and just the noise as a separate
Any help greatly appreciated
When I try and register the noise unless I have the whole bandwith…the HF hiss seems missing…if I select only the offending hiss region, it still doesnt register and if I gain it up, it starts getting very messy
This audio is 00m06s and there isn’t a section without program material (AcGuitar in this case)
I tried with your example, but I don’t find enough noise in there to yeild a useable noiseprint. Your provided separated noise does not resemble the noise of of the program material.
I took a NP from a selection between 3K-17K between program material (1.200s-2.000s) where I quickly rough cut the highs of the guitar out of the selection and registered the NP. Roughly set Tolerance to 14dB; Average unchecked; Ratio to 70%…That probably cuts too much high out for you though, too. There is noise remaining below 3K, but that doesn’t bother me as much as 3-10K does
I originally played in realtime as a guide track as nearly all the tracks for this project are realtime/3 musicians ie no click and I wasn’t monitoring but playing in acoustically hence I didn’t notice the noise ie the record dropped straight in. Then on playback the hiss was obvious so I just recorded noise without the guitar so it should be the same but I’ll double check.
Always hard using passive ribbons on quiet sources…it us going into impedance converter with 20db which loads the tubes nicely but yeah its prob close to 75db gain overall. Im not a fan of SDC and LDC on nylon string…just tinsel
Other than some low end shelf/control it is a -14 target mix so it needs the open space and don’t want anything else in it but I’m thinking I’ll flick some HF on the pre and then I can use denoise a lot more effectively?
the hiss has been changed to sound like filtered noise; the recording is now very honky sounding?
Im just AB the original…which is smooth and full; I might just leave the hiss on…its not distracting but not exactly super clean…but thats just the best that old gear can do.
Thanks but not really what I was hoping for…Id have hoped SL was better than that…i could do what you did simply with ReaFIR…I have to work out another way as…this would be a bit embarrassing to give to a client
Noise removal is always a balance act.. But I think you can greatly improve the result without losing too much.
Perhaps something like this is better than the distracting hiss?
I used “Bertom Denoiser Pro”, one of the best denoisers I know, and it is surprisingly cheap.
If you think this is too much, you can always adjust to taste. And a bit of EQ after denoise always helps.
This would only work if the noise is after the pre.
You can only solve a problem BEFORE the problem source in a recording, meaning: if the hiss is from the Mic, only the source/positioning can make a difference. If it is in the pre, then the source and the mic can be adjusted to get a better signal/noise ratio. if the problem is in the recording medium, then source, mic and pre can make a difference, and so on.
Cheers
Ill check you sample
I already did the best with gain staging etc, proximity already defines the source distance etc but it does have passive eqs on the pre so Im not sure if that will help but I will learn if not.
BTW I mentioned the mic is a passive ribbon…they have no hiss
I did check out Bertom but its essentially a graphic eq which some dynamic control
In reality the SL denoiser is a bit like ReaFIR but leaves serious artifacts if pushed…I think eq would be better in that regard but even 6db would be enough reduction.
Sometimes feels like Im whining…for spectral editing…why isnt this a reasonable task. Its one of the main things I thought SL should be good at right?
Yeah I think this is the most natural… @ctreitzell is close but this seems to have an edge in keeping something.
All helps though…thanks to all who contributed
NOTE
For anyone else looking to solve these type of issues; I got a trial of spectral denoise in Izotop RX11 and it took about 5 mins to learn how to use it…and it is really good…not perfect but more than acceptable
-9db of hiss with no noticeable artifacts etc and I dont mind the slight noise as thats typical of the desk and once gated in the blank sections…
@Aurasphere If you will permit me, this is a blind test . What is your feeling about this result? Does this seem like a good result compared to your existing result with RX11?
Hi
I had as good a listen as I could with earbuds as Im a passenger atm but they are decent.
I think you example is a bit better…the hiss is slightly quieter, very slightly darker in upper mids but its less abrasive than SDN which does not have that in the original.
. What is your feeling about this result? Does this seem like a good result compared to your existing result with RX11?