Coursedesign wrote on 9/7/2006, 9:28 AM
1. It is truly great.

2. It requires a bit of practice before you get it right.

3. There are oodles of posts here with practical tips, just a quick search away.
JeffreyPFisher wrote on 9/7/2006, 9:59 AM
It is a great tool. However, it should be used as the final step after you've exhausted other noise reducing tools. For example, AC can often be attacked with just EQ -- filter out the lows. If that doesn't do it, some expansion or noise gating may suffice. If all that doesn't get you the right sound, then use the Sony NR tool. There is a free d/l, but it has beeps in it to render using it useless.

jlafferty wrote on 9/7/2006, 10:05 AM
I have saved to a text file an echange between DSE and I think JohnMeyer (apologies if I'm off base). Once you see Spot work the plugin in person at one of his seminars, you really get a leg-up.... but the next best thing is this text:

The noise reduction plugin stands by itself. Most advice (in help file and in various tutorials) is to not reduce more than 20dB per pass. Also, pay attention to where you get your noiseprint sample.
williamk wrote on 9/7/2006, 11:32 AM
Used the noise reduction sample download and had an interest experience. Other than beeps which are not in the full version of course.

Ran it through twice at 20 dbs and then listen to it on our computer speaker and it sounded great with almost no noise left over from the AC. Then we plugged in the earphones to listen to it more closely and it sounded like we were on a space ship being attacked by ghosts.. Not sure which noise will be our final output. Will it be the spaceship interview or the nice quiet ghost free interview.

Or are we actually doing something wrong. We tweaked the levels etc to get the great speaker sound.

Dazad and confused and a little haunted.
Jayster wrote on 9/7/2006, 11:41 AM
I'm no expert but it seems to me that wearing a decent set of headphones is always going to be your best bet. Without them, unless your room is near-silent, you'll be bombarded by various other ambient noises, possibly masking what you're trying to eliminate from the recording. I got myself a set of Sennheiser HD-280 noise-reducing headphones and it makes a world of difference. No more listening to my PC fans.

That spaceship sound probably means your noise print was too big of a sample and included some of the "good stuff". The noise reduction is cutting into your real recording.

As was suggested above, zoom in on a portion with no voice content, make a short duration loop, then use it for the noiseprint. If you still get the spaceship sound, delete the noiseprint and do it somewhere else on the timeline.

Hopefully your PC sound card isn't throwing a lot of noise into your monitoring, too. The pros often move the D/A conversion out of the PC, like with a Firewire audio interface. I'm no pro, but I accomplished the same thing by sending SPDIF output from my PC to a set of powered speakers that have a full D/A converting capability.
riredale wrote on 9/7/2006, 11:47 AM
I found NR to be less than wonderful, based on the limited tests I did a few years ago. Yes, you can make background noise appear to go away, but it also takes away the very low-level sounds that weren't noise. On a choral group with piano accompaniment, I thought it made the echos in the hall sound odd. But it could just be me--sometimes we get too sentisitized to certain effects. I'd be curious to see the results of an A-B test with a group.

The spaceship sound (sort of a flanging effect) probably means you need to try a different mode, and reduce the amount of the effect. I recall there was a button you could enable that allowed you to listen to what was removed. If what you're removing sounds natural, then the tweaked audio should sound natural.
Yoyodyne wrote on 9/7/2006, 11:50 AM
Just to chime in - Monitoring audio with small pc speakers and noise concelling headphones could be problematic. They tend to color the sound a lot and when you tweak something to sound good on em' it sounds horrible on anything else.

You can get some cheap near field monitors that should be a much more accurate way to check what your doing. I think KRK and M-audio have some that are well thought of.

Just my .02
Coursedesign wrote on 9/7/2006, 11:50 AM
You can also go nuts trying to monitor on headphones.

Don't EVER do that, unless you are mixing specifically for listening on headphones.

If you don't believe me, get some more experience.
Jayster wrote on 9/7/2006, 11:59 AM
Don't EVER do that, unless you are mixing specifically for listening on headphones. If you don't believe me, get some more experience.

Ok, I don't have a lot of experience, but your comments raise some questions for me. Why do so many studio engineers use headphones? Are the recording studios all amateurs? Or is there something more to it?

Also, filtering out noise isn't the same as mixing. You aren't looking for the quality of the hum sound, you're looking to get rid of it.

Assuming you have decent quality headphones and powered speakers, why not use both? Use the headphones (so you can hear something other than PC fans), do your noise reduction stuff, then listen to it on those powered speakers to see how it really sounds. Then listen to it on your home theater system.
Yoyodyne wrote on 9/7/2006, 12:21 PM
"Why do so many studio engineers use headphones?"

Well for me (and I'm probably barely qaulified to talk about this anyway :) - I use the headphones to listen for extranious sounds when recording and also to check stuff happening in the low freqencies. Things like boom mic handling and low end rumble can sound o.k. on near fields but really bad on larger speakers.

I think the problem with noise reduction and consumer speakers is you are kind of fighting each other. The speakers are going to try and make everything sound "good" so you end up going way overboard with the noise reduction. You then check it on other speakers and it sounds funny so you have to re-tweak for them, and on it goes. I'm not saying you shouldn't use different speakers to check your mix - it's just that using em' for noise reduction (or eq at all really) can be tricky.

This is why nearfields are so handy, they just sound flat - any freqency that is to boosted or too cut really stands out, the general idea being that if you can make it sound good on these things it will sound great on anything else.
farss wrote on 9/7/2006, 3:03 PM
This is very true and really bugs me about a lot of the BS the audio pros sprout about cans. Headphones, even cheap ones are the best way to listen for defects in audio. This is totally, completely and utterly different to mixing. Audio post production isn't mixing for the most part, it's editing and cleaning up. Because headphones are firing the sound straight into your ears your hear everything. all the crud is very easily heard. The only thing to be careful of is you can waste a lot of time cleaning up crud that probably doesn't matter that much unless you plan on adding compression during mixing.

For MIXING, headphones aren't a good look in general, just relying on studio monitors alone isn't a good idea either. Remember most of our audio will be heard through crappola TV speakers, not HiFi systems, check the mix on a TV, in a living room with typical household noise levels as a final test.
auggybendoggy wrote on 9/7/2006, 9:57 PM
I've used both NR2 and currently use soad saop which is more cost effective. HOWEVER you get what you pay for. NR2 works better in my opinion.

SS takes too much quality out of the file and sounds a bit computerized.

My vote goes to sonic Found, but if you dont want to shell it out look at BIAS sound soap lite for a learning experience.

TorS wrote on 9/8/2006, 3:17 AM
-20 dB per pass is way too much. The default is -12, and I think even that is a wee bit drastic. I usually end up around -8 and then make 2 or 3 passes. Do remember to take a new sample each time!
williamk wrote on 9/10/2006, 8:50 AM
Thanks for all the imput. We really like Noise Reduction and it works better with the lower db. We also used some noise gate but not the same effect. Near fields are going to be purchased for our next project and we are going to rely on using varying methods of listening, headphones, computer speakers, sterio speakers, television speakers.

DGrob wrote on 9/10/2006, 11:01 AM
I've used NR2 with Spot's tutorial guidelines very successfully. Lately have been working with a coupla plug-ins (Tone Eliminator, Cleaner, et others) from NewBlue. Actually, pretty good results.


kdm wrote on 9/10/2006, 11:30 AM
re: headphones - yes, they can work well for picking out detail if you don't have a quiet monitoring room - nothing wrong with using them for that as long as they are decent headphones that don't add noise (e.g. ear buds from a laptop headphone jack).

Nearfields and a treated room are best for mixing as they do provide a flatter response, more accurate imaging and depth of field. Headphones can be deceptive in frequency response (esp. low end) and don't provide an accurate stereo image for mixing purposes at least.

re: noise reduction - for another option, Samplitude 9 adds spectral editing which can work well for cleaning up noise. Magix Audio Cleaning Lab has some of this capability, but I haven't used it to say how well it works vs. the options above.

Just a couple of other options in case you decide not to hire an audio engineer to do the audio post work. :-)

Coursedesign wrote on 9/10/2006, 12:53 PM
This is very true and really bugs me about a lot of the BS the audio pros sprout about cans. Headphones, even cheap ones are the best way to listen for defects in audio. This is totally, completely and utterly different to mixing.

Bob, have you ever heard any audio pros say otherwise?

Sometimes when people are busy, they are tempted to give a short answer like "don't use headphones for mixing," which is correct but leaves out the reasons why.

It also doesn't talk about the extra 6-24 hours you may be spending in front of your computer to try to remove the 2,894 flaws that you can hear in your cans but can't hear on regular speakers.

Not to mention that the final "flawless" audio may sound like warmed-over TV dinner (when listened to over speakers) because you mixed for headphones.

epirb wrote on 9/10/2006, 1:26 PM
I second Darryl comments the New Blue filter works very well.
For my needs, along with its other filters its comes with cost wise its hard not hhave it in your tool kit. I experimented (trial'd Sony's) while it is excelent, i can achieve the same results with NB's for the noise I encounter in my projects.
for example I needed to remove room tone from a voice track that i had to rerecord(ADR) so i could replace it with the orig voice tracks room tone. worked great.