Noise control

[RainbowStorm]

I am trying to get more noise control. I tried some pink and white noise reduction tools, technically it eliminated the noise, but I got a much uglier sound. I thought about maybe starting to apply noise control by being more selective with what I record. I would like to have a tool that measures the amount of noise within a signal, for instance within a keyboard sample or after I've put an EQ/compression effect on a track so that I can decide more easily what gear to choose. So far I've been trying to use the statistics function in Nuendo and tried to discover noise by limiting very hard, but I would like to have some tool I can put on the post fader output bus that measures the amount of signal and the amount of noise. Are there any good tools or techniques for this? I tried layering a track of pink and white noise reduction which seemed to eliminate the noise quite effectively, I got a bit too limited sustain though by doing like that so I realised it was a bad idea...

[NeoVXR]

my 2 cents
you may better see all this as a repair job, which is extremely delicate engineering (seek out for "cedar")
noise is basically an issue of recording and production quality. (means also experience and feeling about what to achieve)
cheapo keyboards and bad tape machines, poorly setup guitar amps and stomp pedals, and of course in the opposite keeping every single track as clean as possible.
what can you do if you have a bad source? it's a really specialist job to extract the good cherries from a heap of mud.
the waves restoration bundle is not too bad, when used properly and the material is not totally FUBAR. garbage in garbage out, that it is...

its your question that seduces me to this kind of statement
do you have any specific problem?
there is nothing beyond an experienced ear when it goes to judge negative impact of noise in a commercial production. the type of metering you desire is not known to me, I would call it snake oil analyzing. noise measuring is normally done with test signals and a very incremental procedure. I did not hear of anything that tells you about the "result" of a complex product.
I know what I would do, compare the dynamic, transient and average FFT spectra in a weighted way, and use some neural network to judge (and have it trained by the best audio engineers in the world, who would likely decline the job though). it would remain qeustionable and dependent on music style.

[RainbowStorm]

Originally Posted by NeoVXR

my 2 cents
you may better see all this as a repair job, which is extremely delicate engineering (seek out for "cedar")
noise is basically an issue of recording and production quality. (means also experience and feeling about what to achieve)
cheapo keyboards and bad tape machines, poorly setup guitar amps and stomp pedals, and of course in the opposite keeping every single track as clean as possible.
what can you do if you have a bad source? it's a really specialist job to extract the good cherries from a heap of mud.
the waves restoration bundle is not too bad, when used properly and the material is not totally FUBAR. garbage in garbage out, that it is...

its your question that seduces me to this kind of statement
do you have any specific problem?
there is nothing beyond an experienced ear when it goes to judge negative impact of noise in a commercial production. the type of metering you desire is not known to me, I would call it snake oil analyzing. noise measuring is normally done with test signals and a very incremental procedure. I did not hear of anything that tells you about the "result" of a complex product.
I know what I would do, compare the dynamic, transient and average FFT spectra in a weighted way, and use some neural network to judge (and have it trained by the best audio engineers in the world, who would likely decline the job though). it would remain qeustionable and dependent on music style.

Thanks for your reply! No problems, I just have to do it in my way then... The problem is that all of a sudden I have a quite high idle noise when the drummer counts in. It would be easy to cover up, but I want to target the real cause of the problem. I have a number of suspects. Since I currently have problems with my room's acoustics I haven't been able to recognise noise by ear, so I might have had a mic turned on by mistake during the tracking phase, especially because I had the phantom power option turned on. Another suspect is the drum kit sample in my Yamaha Tyros keyboard or the effects on it. I just have to analyze a bit more, I'm sure I'm able to solve it...

[3rd world order]

Originally Posted by TonyCrazyMan

I am trying to get more noise control.

I never understood why ppl are so anal about noise...

[Reptil]

euh well I was asked if I could clean up some recording of salsa lessons, recorded with the cheapest wireless gear available. I could hardly hear the guy give his salsa instructions because of a loud hum from the mic, some more from the wireless, a hiss from the cheap preamps in the cheap mixer, and muck from the md recorder. and it was badly set up as well. I guess you can call that noise. Garbage in garbage out.

[NeoVXR]

MD means ATRAC i suppose?
lossy compression makes denoising much worse in my experience...
at least there should be good moments in the recording to capture footprints.

[Reptil]

yeah I did get footprints, then tried to get rid of it. with software (TC and bias)
no dice, it took away an equal portion of the voice. rendered the rest very unnatural.
then I tried a little eq on my desk (very soft) and some on a parametric. this helped a little bit, in clearing things up. but not much. Then I tried a gate, before and after eq. very slow, but this also didn't help much. As a last resort I tried different settings on the eqs, parametric in front, then decompression on my Kultube, and after that some soft eq on the desk. also I added some volume changes in the software.

aaarhgg still sounded like cr*p
sometimes there is no rescue possible
I gave the file back, on an usb stick, no charge
guess the guy wasted some money in getting the whole ponyshow there, and then trying to cheat on quality. (this has been common practice in this country for countless centuries LOL) the audio quality part is mostly overlooked in "multimedia" projects.
it was supposed to be the soundtrack for an educational video.

yes atrac minidisc bleuh hate the whole format.
a lot of "artists" in this country bought one some time ago, to "record" ideas.
also you can find the "pro" model from Sony in a lot of clubs. (to replace DATs)
I wish they didn't. It makes the whole experience so much less enjoyable....

and oh yeah the guy was completely surprised, he thouht it would be ok, because it is "digital". gawd talking about stone age attitudes. sorry end of rant...

[joeq]

Originally Posted by TonyCrazyMan

I thought about maybe starting to apply noise control by being more selective with what I record. .

that's the key. Be like my grandmother when she is on Light Patrol- switching off unused lamps throughout the house. It may seem petty, but those pennies add up.

Noise reduction software should come in a glass case with a little metal hammer and a sign that says "In case of emergency, break glass"