Topic: Opinions about post-processing - how much is too much?

[robgronotte]

I've been working extensively on my master recordings recently (mostly with iZotope RX10), trying my best to get rid of anything that I consider flaws in the recording, including removing things that were actually heard by the audience, such as annoying talkers, whistlers, and screamers.
In a few cases I have also removed sounds that came from the stage and/or PA, mostly annoying feedback which clearly wasn't intentional. 

All of these do somewhat detract from the recording as a document of a particular show, but I feel it is definitely worth it to improve the listening experience.

However, I am sure that some people might think what I have done is too much, and I would like to know others' opinions about things that might come up in my editing.

I'm currently working on a recording I made where the audience was mostly very quiet, and the sound was very good (but quiet), so there's not a lot that needs to be done apart from reducing the volume of the applause between songs and doing my best to reduce the sounds of some audible coughs, as my gf and I were both fighting a cold at the time.

However, during some of the songs, there are quite a few instances of fret squeak when the acoustic guitar player changes chords.  I'm torn about whether or not to remove this.  I could likely do it pretty cleanly, and I'm fairly certain this noise was not something the artist deliberately wanted to be heard at the concert, but it still feels more revisionist than most other editing I have done.  Removing it would definitely make the recording more listenable to me, and I'm sure to most others as well, but it is completely changing something which was heard by every member of the audience at the show.

I would love to hear people's thoughts about removing the fret squeak, as well as more general thoughts about changing things in post-processing.

[daze]

I've always been in the "less is more" camp for post-processing, although I'm trying to learn better izotope skills to deal with mostly annoying obtrusive audience noises/issues.  I'd also like to hone my EQ skills (and learn Ozone as I have a few muddy recordings that can use a little brightening). 

As far as fret squeak noises, you could always save one version before you do that, and then another one after (having the best of both worlds).  And concerning other PA issues (unintentional noises), the only time I remove that is if it's really loud and egregious. 

[Fatah Ruark (aka MIKE B)]

I think with things like EQ and compression it's important to rest your ears. Do what you think sounds good, and go back and listen to it again in a day or so. I find that I frequently EQ or compress a bit too much and then have to back it off a bit before sharing. With time I've gotten better at this, but still like to take a break prior to calling it good.

[daze]

. . .
I'm currently working on a recording I made where the audience was mostly very quiet, and the sound was very good (but quiet), so there's not a lot that needs to be done apart from reducing the volume of the applause between songs . . . .
. . .

Concerning reducing the applause volume between songs, if it's not embedded in artist chat/music, I use the envelope function to bring it down, but I continue to struggle where the loud between-song applause/screams bleed into the artist chat/music.  If anyone has any tips on dealing with that, thanks in advance.

[robgronotte]

. . .
I'm currently working on a recording I made where the audience was mostly very quiet, and the sound was very good (but quiet), so there's not a lot that needs to be done apart from reducing the volume of the applause between songs . . . .
. . .

Concerning reducing the applause volume between songs, if it's not embedded in artist chat/music, I use the envelope function to bring it down, but I continue to struggle where the loud between-song applause/screams bleed into the artist chat/music.  If anyone has any tips on dealing with that, thanks in advance.

In this recording the audience was very respectful of the quiet venue (formerly a church) so made almost no noise during the music, and very little when the singer was speaking. I find this is generally the case for shows that are so quiet that the cheering between songs is louder than the music.

I always run once through the iZotope de-click module before doing anything else, which lowers the loudest nearby claps.  It's going to be tough to reduce other applause over the music, but you could get rid of random woos and whistles with the spectral repair (currently discussed in another thread).

The same thing for woos and whistles over on-stage talking.  I haven't run into the problem of loud applause over on-stage chat, but you could try separating the speech from the other noises and then boosting the speech.  I find UVR5 is great overall for isolating speech, but it does sometimes have trouble with speech under loud applause.

[daze]

Thanks!  I'll give UVR5 a whirl.  Unfortunately I have a lot of shows with the embedded applause/screaming, and I'm willing to give anything a try!

[roffels]

I try to find a balance.

The more "perfect" a recording is, counterintuitively I'll spend more time cleaning the sound up. If it's a quiet acoustic show and a good board and audience recording, I'm more likely to try to clean up in RX any time someone drops a soda can or coughs or their chair squeaks, etc. I've found you can get away with a lot of cleaning when you're working with multiple sources and mixing them together.

And maybe it's sacrilege, but if there's a faulty cable that causes a loud spike in noise, or some nasty feedback, I'll try to attenuate those. I'm not trying to present a perfect recreation of the event - that's impossible because no one was standing where my microphones were, my microphones don't perfectly replicate the human ear, and everyone there had a different experience. I just want to make a good sounding tape.

If it's a rowdy crowd, I just let it be. 

[roffels]

Concerning reducing the applause volume between songs, if it's not embedded in artist chat/music, I use the envelope function to bring it down, but I continue to struggle where the loud between-song applause/screams bleed into the artist chat/music.  If anyone has any tips on dealing with that, thanks in advance.

In Izotope RX, you can visually identify the loud parts and attenuate the specific frequencies that may or may not overlap with the speaking. So if there's a loud "Woo", zero in on that and lower the volume a bit.

[daze]

Concerning reducing the applause volume between songs, if it's not embedded in artist chat/music, I use the envelope function to bring it down, but I continue to struggle where the loud between-song applause/screams bleed into the artist chat/music.  If anyone has any tips on dealing with that, thanks in advance.

In Izotope RX, you can visually identify the loud parts and attenuate the specific frequencies that may or may not overlap with the speaking. So if there's a loud "Woo", zero in on that and lower the volume a bit.

Thanks.  That works sometimes, but I'm trying to figure out the situations where there are layers and layers of embedded applause/screams.  I know I won't be able to totally eliminate them but I'd like to at least be able to reduce them.

[robgronotte]

Concerning reducing the applause volume between songs, if it's not embedded in artist chat/music, I use the envelope function to bring it down, but I continue to struggle where the loud between-song applause/screams bleed into the artist chat/music.  If anyone has any tips on dealing with that, thanks in advance.

In Izotope RX, you can visually identify the loud parts and attenuate the specific frequencies that may or may not overlap with the speaking. So if there's a loud "Woo", zero in on that and lower the volume a bit.

Thanks.  That works sometimes, but I'm trying to figure out the situations where there are layers and layers of embedded applause/screams.  I know I won't be able to totally eliminate them but I'd like to at least be able to reduce them.

Unfortunately you won't likely be able to separate the screams from the onstage chat with UVR5, as they're both human voices.

[roffels]

Concerning reducing the applause volume between songs, if it's not embedded in artist chat/music, I use the envelope function to bring it down, but I continue to struggle where the loud between-song applause/screams bleed into the artist chat/music.  If anyone has any tips on dealing with that, thanks in advance.

In Izotope RX, you can visually identify the loud parts and attenuate the specific frequencies that may or may not overlap with the speaking. So if there's a loud "Woo", zero in on that and lower the volume a bit.

Thanks.  That works sometimes, but I'm trying to figure out the situations where there are layers and layers of embedded applause/screams.  I know I won't be able to totally eliminate them but I'd like to at least be able to reduce them.

I'm guessing you're talking about an audience only tape? That would be quite the challenge. Most of my tapes are board+aud or multitrack+aud so I can get away with a lot more in those situations.

[opsopcopolis]

I'll generally go all out with set and forget processing like EQ and comp/limiting. I never have and never will do any manual noise reduction of audience noise, mainly because I don't have the time and don't really care enough.

[Chanher]

I think with things like EQ and compression it's important to rest your ears. Do what you think sounds good, and go back and listen to it again in a day or so. I find that I frequently EQ or compress a bit too much and then have to back it off a bit before sharing. With time I've gotten better at this, but still like to take a break prior to calling it good.

This is 100% true for me. EVERYTIME I finish a long mastering session, I save it and come back later because there’s always something I can improve using fresh ears. Repeat this cycle until it’s the best you can get it, and don’t hesitate to listen on an alternate playback system. I like to try a shitty bluetooth speaker, the car, and even ear buds. These micro-adjustments can mean the difference between a good and great final product.