19th December 2012
|
#1 | | Gear Head
Joined: Jan 2011
Posts: 57
Thread Starter | Removing Hiss
Hello guys, little noise problem ...
Working on Cubase 5.5.3 on Windows XP.
Loaded up some wav files recorded for me elsewhere. Three of the guitar tracks have a hiss type noise at the end as the last note decays. Although there are some other instances, it's only really apparent as the last chord decays in the finished mix. So what's the best remedy? Do I run the mix through in Wavelab or do I reprocess the offending individual tracks and re-export? The only plug in I have is Waves X Noise ...
I thank you all in advance.
Axxe
|
| |
19th December 2012
|
#2 | | Lives for gear
Joined: May 2012
Posts: 1,208
|
You could add a digital delay effect to the decaying note, which will mask the hiss and not add any noise itself.
|
| |
19th December 2012
|
#3 | | Gear Head
Joined: Jan 2011
Posts: 57
Thread Starter | Quote:
Originally Posted by OpusOfTrolls You could add a digital delay effect to the decaying note, which will mask the hiss and not add any noise itself. | Nice idea but not an option in this case...
I guess what I am wondering is whether running three tracks through the plug in is better than running the whole mix through once?
|
| |
19th December 2012
|
#4 | | Lives for gear
Joined: Jan 2009 Location: Boise, Idaho
Posts: 2,578
|
What I do in cases like this is make a copy of JUST the offending portion (with a little pad on either end) of the offending track and process that JUST enough to make it not annoying. I then inter-cut that with the unprocessed version so there's only NR in the spots that need it and only on the tracks that need it.
X Noise isn't very good, so you'll have to be super careful. Noise reduction in general is best avoided, so you might also try an EQ tuned to reduce the noise with as little affect on the rest of the audio as possible. You can cross-fade the decaying noisy track with the EQed version. If it's not too bad, sometimes I just shorten the fade a little.
|
| |
19th December 2012
|
#5 | | Lives for gear
Joined: Dec 2007
Posts: 875
Verified Member | Quote:
Originally Posted by axxeman I guess what I am wondering is whether running three tracks through the plug in is better than running the whole mix through once? | Running every offending track through noise reduction is better than running a whole mix.
|
| |
19th December 2012
|
#6 | | Lives for gear
Joined: Jan 2007 Location: Minneapolis
Posts: 1,504
Verified Member | Quote:
Originally Posted by Alexey Lukin Running every offending track through noise reduction is better than running a whole mix. | Definitely. That allows you to fine tune the processing per track while leaving the good tracks alone. Running the whole mix through should be left as a last resort.
|
| |
19th December 2012
|
#7 | | Lives for gear
Joined: Jan 2010 Location: underneath the dank, cobbled streets of Landon Taaaan'
Posts: 1,858
Verified Member | Quote:
Originally Posted by Greg Reierson Definitely. That allows you to fine tune the processing per track while leaving the good tracks alone. Running the whole mix through should be left as a last resort. | Yes! Also, can consider running reduction on the offending section only. I quite often will use a little reduction on a quiet intro or outro and leave the main sections alone, as the noise is usually adequately masked.
|
| |
20th December 2012
|
#8 | | Gear Head
Joined: Jan 2011
Posts: 57
Thread Starter | Quote:
Originally Posted by Greg Reierson Definitely. That allows you to fine tune the processing per track while leaving the good tracks alone. Running the whole mix through should be left as a last resort. | Yes, I think that I didn't really consider that at first. Probably looking for the one step fix. The whole mix idea went out of the window when I realised that the crash cymbal on the last note lost its sizzle when processed, (as, of course, it would). So I'll try running the three offending tracks through separately and then recombining.
Oddly enough, there is no hiss pre intro but at the end it sounds like a VCR recording where the automatic level control is kicking in on a quiet passage in a film. Bloody annoying.
If anyone has settings suggestions for that plug, I'd welcome them. Alternatively, is there a better tool I could get as a demo version to do the job?
Cheers people.
|
| |
20th December 2012
|
#9 | | Gear nut
Joined: Apr 2007
Posts: 120
|
You could try demoing iZotope RX2 maybe?
|
| |
20th December 2012
|
#10 | | Gear maniac
Joined: Sep 2012
Posts: 186
|
TC Backdrop
|
| |
20th December 2012
|
#11 | | Gear nut
Joined: May 2007
Posts: 132
|
Rx2 is incredible. Just get the demo trial. Or you could just pretend and say its tape hiss. Then you'd accept it  waarmm
|
| |
20th December 2012
|
#12 | | Gear addict
Joined: Feb 2010
Posts: 371
|
go gentle with the noise reduction.
It is tempting to try and eliminate all the noise in one shot [ > -10dB] ... however, this can cause other sonic issue. Best to go in small reductions.
Rx is a very useful tool. I rely heavily on it for my restoration work.
|
| |
20th December 2012
|
#13 | | Gear Head
Joined: Jan 2011
Posts: 57
Thread Starter | Quote:
Originally Posted by paintitblack Rx2 is incredible. Just get the demo trial. Or you could just pretend and say its tape hiss. Then you'd accept it  waarmm | Is it simple to use out of the box? The website makes it seem complex at first glance. I looked at Voxengo Redunoise but seems that the trial won't work unhindered in the same way as Izo's.
To justify the pricetag on the izotope RX, I'd have to have a LOT of tracks needing treatment rather than (at this moment) just one.
|
| |
20th December 2012
|
#14 | | Lives for gear
Joined: Jan 2009 Location: Boise, Idaho
Posts: 2,578
|
RX is actually really intuitive if you're already familiar with digital noise reduction in general. I just got the demo and really like it.
|
| |
20th December 2012
|
#15 | | Gear Head
Joined: Jan 2011
Posts: 57
Thread Starter | Quote:
Originally Posted by wado1942 RX is actually really intuitive if you're already familiar with digital noise reduction in general. I just got the demo and really like it. | Thanks. I'm dowmloading it now. The RX2, not the advanced. I wouldn't imagine I need anything more than the basic set up for this job.
I guess that it is best to run each track separately through wavelab and then re-import into the project, rather than clog up the cpu with a heavy workload? If you do render the single tracks as I have suggested and then re-import, is there any latency time shift or is the audio in the same place as before? Probably a stupid question but I gotta ask it |
| |
20th December 2012
|
#16 | | Lives for gear
Joined: Jan 2009 Location: Boise, Idaho
Posts: 2,578
|
I have the advanced version. I haven't looked at latency really. Like I said earlier, what I do is make a copy of the portion that needs treatment and treat JUST that portion. When I reimport it, I just manually match it to the original track.
|
| |
20th December 2012
|
#17 | | Gear addict
Joined: Feb 2010
Posts: 371
|
I use RX2 more often as a stand alone app.
|
| |
21st December 2012
|
#18 | | Gear Head
Joined: Jan 2011
Posts: 57
Thread Starter | Quote:
Originally Posted by RJHollins I use RX2 more often as a stand alone app. | I don't think the trial version will let you do that? Save disabled??
|
| |
21st December 2012
|
#19 | | Gear Head
Joined: Jan 2011
Posts: 57
Thread Starter | Quote:
Originally Posted by wado1942 I have the advanced version. I haven't looked at latency really. Like I said earlier, what I do is make a copy of the portion that needs treatment and treat JUST that portion. When I reimport it, I just manually match it to the original track. | Hmm, so maybe I could slice the three tracks at the start of the final note/chord, process just these 'tails' (as the hiss is inaubible elsewhere in the mix), and then 'glue' them back from whence they came?
Or I email the offending wavs to some kind contributor and ask nicely |
| |
21st December 2012
|
#20 | | Gear addict
Joined: Jun 2011 Location: PA
Posts: 364
|
Depending on how long and prevalent your tails are - you might be able to have a gate close just a little ahead of your hissing. Do you really need to hear that chord tail past -40?
Depending on how solidly stable your hiss is you might be able to take hiss only / inverse phase and slide under your tails sum into a sub-mix for each instance of a track.
If you can not get a trial of RX2, I think you can get a trial of Audition CS6, same concept of spectral view editing, mess with it and see how it treats you.
|
| |
24th December 2012
|
#21 | | Gear Head
Joined: Jan 2011
Posts: 57
Thread Starter | Quote:
Originally Posted by RRCHON Depending on how long and prevalent your tails are - you might be able to have a gate close just a little ahead of your hissing. Do you really need to hear that chord tail past -40? Depending on how solidly stable your hiss is you might be able to take hiss only / inverse phase and slide under your tails sum into a sub-mix for each instance of a track.
If you can not get a trial of RX2, I think you can get a trial of Audition CS6, same concept of spectral view editing, mess with it and see how it treats you. | Now THAT is spooky. I hadn't checked back to this thread since my last post and, whilst in the shower half hour ago, I was wondering about using the hiss as a separate track and mixing it in with phase inversion to cancel itself out. Hopefully, the results would be pretty OK and leave the instruments (largely) sonically undisturbed?
Is it 90 deg or 180 for phase cancellation?
Other concept I considered was running the track(s) through a high pass filter set to just below the hiss frquency, then invert this and mix in. Finally resynthesise the top end with some gentle excitation ... Could be total bollocks though :(
Cheers.
|
| |
24th December 2012
|
#22 | | Lives for gear
Joined: Jan 2009 Location: Boise, Idaho
Posts: 2,578
|
Hiss is a random noise. If you sample hiss and mix it back with the noisy track, it doesn't matter if the polarity is matched or inverted, you will always wind up boosting the hiss by 3dB. A trick like that might work with something non-random like AC hum, though.
|
| |
24th December 2012
|
#23 | | Lives for gear
Joined: Apr 2011 Location: Vermont
Posts: 2,241
|
i've used dmg compassion (expander/filter sections) for noise reduction/removal
you can use it sort of like a 'multi-band' and 'split' off the freq. range you want to process and leave the rest untouched...
works good
__________________
"it's a long way to the top if you wanna rock -n- roll" -Jesus
|
| |
24th December 2012
|
#24 | | Gear Head
Joined: Jan 2011
Posts: 57
Thread Starter | Quote:
Originally Posted by wado1942 Hiss is a random noise. If you sample hiss and mix it back with the noisy track, it doesn't matter if the polarity is matched or inverted, you will always wind up boosting the hiss by 3dB. A trick like that might work with something non-random like AC hum, though. | What, even if you are just inverting the mirror image of the offending material and aligning it timewise? Obviousy, a little hands on tinkering is the way to go but it's all closed off till Jan thanks to Xmas intervening.
|
| |
24th December 2012
|
#25 | | Lives for gear
Joined: Feb 2003
Posts: 4,785
Verified Member | Quote:
Originally Posted by axxeman What, even if you are just inverting the mirror image of the offending material and aligning it timewise? Obviousy, a little hands on tinkering is the way to go but it's all closed off till Jan thanks to Xmas intervening. | If there is source material with that hiss that you are trying to preserve it will be affected also. If there is no source that you really need why not just put a volume automation on it?
The hiss is probably caused by compression releasing on the guitars.
I have treated similar problems by stretching the guitars ending notes and then putting a volume slope on the to fall off before the hiss becomes audible. Adding a little reverb tail can also buy you some time to fade.
|
| |
25th December 2012
|
#26 | | Gear Head
Joined: Jan 2011
Posts: 57
Thread Starter | Quote:
Originally Posted by Rick Sutton If there is source material with that hiss that you are trying to preserve it will be affected also. If there is no source that you really need why not just put a volume automation on it?
The hiss is probably caused by compression releasing on the guitars.
I have treated similar problems by stretching the guitars ending notes and then putting a volume slope on the to fall off before the hiss becomes audible. Adding a little reverb tail can also buy you some time to fade. | I didn't explain myself correctly. After the audio ends there is 200 - 300 ms of pure hiss only. I was proposing to copy and paste this onto a new track and append it up to the required length, then invert the phase and mix in.
|
| |
25th December 2012
|
#27 | | Lives for gear
Joined: Dec 2007
Posts: 875
Verified Member |
It will only work for those 200–300 ms, for the reasons explained by wado1942. Generally, phase-inverting a noise sample is pointless: you can achieve similar or better results with a gate.
|
| |
25th December 2012
|
#28 | | Gear Head
Joined: May 2005
Posts: 43
|
Waves x-noise is perfect for this. You already have the perfect tool for what you're trying to do. I use x-noise all the time to solve this exact problem. I've used it on hundreds of guitar tracks to get rid of noisy amps, him, hiss, etc.
I find the clearest spot where there is only the noise I'm trying to get rid of and sample it in x-noise. A lot of times at the beginning of the song where the player has the volume up but hasn't started playing yet, or during a short break, etc. That's the key, getting a good sample of just the noise itself first. Then just crank up the threshold and reduction to your liking and you're done. Automate the plugin to only turn on during the problem spots. This takes all of about 2 minutes to accomplish and the results are exactly what you're looking for... no noise and no evidence of artifacts/processing.
Sure there are other tools out there to do the job, but why look for fifty other solutions when you already have a great one right in your pocket.
|
| |
25th December 2012
|
#29 | | Gear Head
Joined: Jan 2011
Posts: 57
Thread Starter | Quote:
Originally Posted by rainmaker Waves x-noise is perfect for this. You already have the perfect tool for what you're trying to do. I use x-noise all the time to solve this exact problem. I've used it on hundreds of guitar tracks to get rid of noisy amps, him, hiss, etc.
I find the clearest spot where there is only the noise I'm trying to get rid of and sample it in x-noise. A lot of times at the beginning of the song where the player has the volume up but hasn't started playing yet, or during a short break, etc. That's the key, getting a good sample of just the noise itself first. Then just crank up the threshold and reduction to your liking and you're done. Automate the plugin to only turn on during the problem spots. This takes all of about 2 minutes to accomplish and the results are exactly what you're looking for... no noise and no evidence of artifacts/processing.
Sure there are other tools out there to do the job, but why look for fifty other solutions when you already have a great one right in your pocket. | Walk me through the process please ...
|
| |
25th December 2012
|
#30 | | Gear Head
Joined: Jan 2011
Posts: 57
Thread Starter | Quote:
Originally Posted by Alexey Lukin It will only work for those 200–300 ms, for the reasons explained by wado1942. Generally, phase-inverting a noise sample is pointless: you can achieve similar or better results with a gate. | Understood and, thanks.
|
| | | |