Topic: Is there software that removes cellph. RFI/ Is my recording toast?(w-whole show)

[igene]

My fault. My friend kept texting me for directions to the show while I was taping.
I had to respond. It happened 5 or 6 times during the recording.
He made it for the last 3 songs 

The recording is amazing, the interference is really obnoxious.
It ruins the recording 

Help !!!

It would be a nightmare doing this manually.

[Fatah Ruark (aka MIKE B)]

You're SOL. No way to remove it.

Better find out what part of your rig is allowing the interference to get into the recording, otherwise this will keep happening. I would guess a cable first.

I have never had this issue (knock on wood).

EDIT: I would also show your friend this website:

http://maps.google.com

Very handy!

[igene]

very funny - google maps...
He had a really old type phone at the time (has an Android now) and was calling from the road.

I have SP-CMC-08 mics, and a Blackberry cellphone..
Bad combination apparently. I now know they put out serious noise.
I always turn off the phone now at shows.
This happened last summer @ !!! in Brooklyn, out side show.


I just DL'ed a !!! show from TD and realized my recording is better.
I figured if I could fix the problem, I'd post it.

It's too damn bad, the recording is truly VG.
It was the 1st show with those mics. Who knew.

No more cellphone while stealthing ! period.
(and no, I don't record to the BB)

[page]

Depends on the type of interference.  Short clicks, you can zoom in on the clicks and get them out with a very quick volume envelope.  Or try a declicking plugin.  If that doesn't do the trick, try a spectral editor.  Of course any interference that is at a given frequency you can kill with EQ, but I am guessing this is loud broadband interference that is not periodic.  That can be pretty tough . . .

Yeah, I was going to suggest Izotope's RX. If anything can eliminate short bursts its that (once you get past the learning curve). It will still be a manual and laborious process though.

[setboy]

I have had the same thing happen to me with my SP-CMC-08. probably a bad cable connection or something

[igene]

They were BRAND NEW at the time.

Since then, I don't have the phone on, and there are no problems.
I see folks using phones near me, but I'm not affected.

As a matter of fact, I've made some of the best stealth recording with this rig.
CMC-08>CA-9100>VR-10

Thanks for the Izotope's RX suggestion.
I'm looking into it right now.
I've never heard of it, so 

I'll do a sample and post it in a few so you can hear the issue.

[igene]

here the sample.
I converted it to mp3 / 320Kbs for a smaller file.
The original is wave/16bit/44.1

it's not a complete song, just a partial with the noise and then a bit more for reference.

!!! 2010-08-15 Williamsburg Waterfront, BK, NY -cell noise sample
http://www.megaupload.com/?d=NRNK2MS4

[trustthex]

You always could use the other source for patches, if it cannot be repaired.


That's blackberry interference for sure, good luck!

[igene]

no other source. the other !!! show was another show, another location.
I never see them posted, so I DL'ed it to hear the quality.

My overall sound quality is better, but I won't post it with that horrible RFI.

I guess I'm boned !

I downloaded a free copy of Izotope's RX advanced.
It doesn't look like it covers RFI.

I don't think anyone does.
(I hope I'm wrong)

[Brian Skalinder]

In general, I think a flawed recording is better than none.  Sometimes not, though.  If the recording and performance are otherwise worthwhile, and no other recording exists, you could <1> minimize the negative impact to the listening experience by reducing the volume and intensity of the RFI, or <2> cut out the RFI and crossfade to minimize the impact, and post anyway.  That's probably what I'd do.

[Fatah Ruark (aka MIKE B)]

Great band.

If it's only in 5 or 6 spots I say just share it. Over an entire show that isn't a huge deal.

[igene]

did you listen to the sample ?

That happens 5 or 6 times through the show, but it lasts for like 20  or more seconds. each time.
and then repeats the cycle soon after.
there are many songs w/o it too.

I guess I'll try to fix it, but that's a bear of a job.

There's so much noise present in each instance of the RFI, it's a little overwhelming.

[page]

I downloaded a free copy of Izotope's RX advanced.
It doesn't look like it covers RFI.

I guess I'll try to fix it, but that's a bear of a job.

yes, you have to do it by hand in RX, there is no plugin to remove it. The reason I suggested RX is because it gives you a spectral view and allows the selection of sounds and harmonics in one shot.

I'll look at the sample when I get home and see if I can tinker with it in RX.

[Gutbucket]

The tools in RX are amazing (I haven't used RX but have played around with similar tools in Samplitude's Master Edition) but take both patience and skill to use effectively.

There's so much noise present in each instance of the RFI, it's a little overwhelming.[/i]

It may be helpful to consider that usually the most sucessful approach to noise-reduction is reduction, not total elimination.

My suggestion from a practical standpoint- if reducing the noise doesn't work (I haven't listened but I think it will be difficult) consider using a simple volume envelope to fade the program level down significantly just befor the events occur, then back up to normal afterwards.   The music will be reduced in level during these events as well of course, but at least the noise won't be shockingly loud, the fade down just before it occurs will give the listener a short 'warning' each time, and yet the flow of the music won't be completely broken as if there were a total dropout or gap.

[igene]

what's funny is I think the noise is AT the same level as the music, so that technique will not be helpful.
I appreciate the input though.

I use Goldwave, and I'm really good with it.

If I use that Software, and reduce each instance at the sine wave level, I think I'll go insane.
I've done minor fixes of DAT digi noise. That usually only happens 1 or 2 times and may involve actual program removal
and matching to the next good section of audio.
Sometimes matching time and cutting out a whole chorus to have it sound good etc...

This noise here is so interwoven into the music, it would take a solid week of work to fix.
I'm not sure I'm ready for that.

I'm uploading the show right now to MEGAUPLOAD and I'll post the link here.
I'll let you guys decide if I should start this insane project.
You basically should hear the whole show to put the RF noise into perspective.

[Gutbucket]

what's funny is I think the noise is AT the same level as the music, so that technique will not be helpful.
I appreciate the input though.

Sure, I realize that. What I'm suggesting is basically the same as manually turning the level down using the volume knob each time the noise occurs.  The only difference is that by using volume automation envelopes you do that automatically for the listener and make for a better listening experience even though the noise is still there.

[igene]

Ok.
here it is.
I'll check back in later, or tomorrow to check the feedback.

http://www.megaupload.com/?d=NOFTH8ZL

[page]

Now that I've listened and played with it, my verdict is still to use RX. I just about vanquished the first machine gun blast in about 10 minutes of fiddling with it (including the time it took to play with settings). You won't eliminate it unless you take the time to really dig in and work with the individual segments, but you'll definitely be able to put a serious dent in it either way without losing a chunk of your life in the process. I actually spent more time doing the screenshots and typing all this up. 


Before






After






Again, you won't get rid of all of it unless you put in a good 7-8 hrs of work, but you can definitely minimize it in about 2-3. The trick here is since it's a constant short bursts, treat them as clicks and use the declicker/decrackle tool. It didn't like the random clicks nearly as much as the predictable periodic ones. I did it as one big swath, but you'll do better to select sections (like the first half of a blast, and then the second half of the blast separately). Also, while I used a strength of 10 to illustrate, it sounded ok, but I recommend something closer in the 5-8 range to keep the audio from turning to mush. I find it's better to under-do then over-do in this arena. I didn't even try the decrackle tool, figured the above was a nice enough illustration.

I reckon a good day (or maybe just an afternoon) to learn RX well enough, then another 1-2 hrs playing with the settings to really maximize your benefit, and another actual 2-3 in fixing the entire show. At the end of it, you're out that time, but you'll be able to fix all sorts of stuff later without having to relearn the product. This is actually easier in terms of labor and time than the kickdrum example I mentioned earlier in that the filter works with you. Best of luck!

[igene]

Looks like I've got some learning ahead of me.
I gave it a try, and I can tell this is going to take some time.
Thanks for the advice.
Going to start hacking away at this later today.

Hope it comes out sounding normal, not processed.

[page]

Looks like I've got some learning ahead of me.
I gave it a try, and I can tell this is going to take some time.
Thanks for the advice.
Going to start hacking away at this later today.

Hope it comes out sounding normal, not processed.

That's part of why I suggested backing off from a value of 10 to say 5 or 8. 

But anyway, play around with all of those tools, try selecting different amounts of the noise (large sections, small sections, large sections with just small bits of noise, etc). Yeah, RX has a learning curve, but it's a nice tool to have in your arsenal anyway. Good luck.

[igene]

thanks.

I''ll post something when I have progress.

[TNJazz]

I say leave that shit in there.  When people bitch about it just post that this is what cellphone interference does to a recording.  Turn your phones off!

Call it a PSA.

[igene]

did you listen to the recording ?

[Fatah Ruark (aka MIKE B)]

Good luck. Hopefully you can salvage it.

!!! is a great band.

[keytohwy]

My fault. My friend kept texting me for directions to the show while I was taping.
I had to respond. It happened 5 or 6 times during the recording.
He made it for the last 3 songs 

The recording is amazing, the interference is really obnoxious.
It ruins the recording 

Help !!!

It would be a nightmare doing this manually.

I downloaded the sample.  Is there a relatively quiet part with the noise in it?  If you, so you may be able to isolate it and remove from the other parts.

[runonce]

My fault. My friend kept texting me for directions to the show while I was taping.
I had to respond. It happened 5 or 6 times during the recording.
He made it for the last 3 songs 

The recording is amazing, the interference is really obnoxious.
It ruins the recording 

Help !!!

It would be a nightmare doing this manually.

I downloaded the sample.  Is there a relatively quiet part with the noise in it?  If you, so you may be able to isolate it and remove from the other parts.

Audacitys noise reduction tool used to work pretty well - but only if you have a music free sample of the noise...

If the problem is recreate-able - maybe record some silence + noise - and use that for your noise profile in Audacity.

[igene]

So in a quiet room, run the recorder with good levels, and text a few times ?
then take that cell noise w/o any background noise and use it as a noise sample in Audacity ?

I've never used that software.
 Does it pull out the noise it is given as a sample or something ?

I should get this, sounds easier then the previous way described.

please explain further !

[johnw]

is there a trial version of Rx? I was thinking that software was like $3-400.

[TNJazz]

did you listen to the recording ?

yes.  it is what it is.  unless you got paid for that gig and have to deliver to a client/customer, I say move on and focus on the next one.

I have a couple of recordings I made over the years that have the same problem.  Shit happens sometimes.  Can you fix it?  Maybe.  Is it worth the time and effort?  Well...personally I don't have hours of time to spend on audio restoration unless someone is paying for my time.  You may feel differently of course.  Just offering my opinion.

[igene]

Well, I agree and disagree.

let me explain.
1. No, I don't have time right now to work on it for endless hours
2. No, I'm not getting paid for the gig/recording - so it's not that vital
3. However, if I learn the software now, and can get good with it (assuming it does a good job)
    if I come across this issue in the future  - I'll know how to fix it quick.

I've had to deal with this before. Fortunately I had a crappier 2nd audio source to work with
(video cam audio) and I fudged it a bit and it flew. whew ! client happy !

so I'll get back to this soon (as a learning experience)
It would be nice to hear this recording w/o the bullshit RFI all over it too.
It's really not a bad recording.