Become a Site Supporter and Never see Ads again!

Author Topic: Post-Processing - Corrective EQ on AUD-Recordings  (Read 615 times)

0 Members and 1 Guest are viewing this topic.

Offline Pittylabelle

  • Trade Count: (0)
  • Taperssection Newbie
  • *
  • Posts: 31
  • Gender: Male
Post-Processing - Corrective EQ on AUD-Recordings
« on: May 07, 2018, 05:41:58 PM »
Can someone please give me some tips on Corrective EQ?

Basically, my OKM Rock Microphone tends to show a not quite linear progression - regardless of the location and the PA. And in the range 4500-8500 Hz it also falls off a bit (please see screenshot). When I try to "straighten" the EQ course in iZotope RX, the recording sounds noticeably better. I've put together a few screenshots of "iZotope RX", as a basis for discussion.

1. Typical frequency course of a professional recording.jpg (In principle all EQs follow the course of this example, wich is also similar to Pink Noise)

2. Pink Noise course example.jpg

3. Typical frequency course of my own recordings.jpg

https://drive.google.com/drive/folders/1FOIlXIOcXJ5TYbSfU_wfQDnwTISmlab3?usp=sharing

Now to the actual problem:
If I increase certain frequency ranges in my recording in such a way that a linear progression occurs, clipping inevitably occurs. On the other hand, if I say to the program that it bluntly takes over the EQ of a professional "similar" recording, the new course of my original recording is nearly similar, but no clipping occurs.

However, I just would like to get the basic characteristics of my original recording and just correct the frequency response a bit.

So what would be the correct course of action, starting from my RAW recording?

Option 1:
Raise the frequency areas with lower ranges. As a result, clipping occurs.

Option 2:
Reduce the overall level first, to get enough headroom so that no clipping can occur. Then raise the frequency areas with lower ranges. Then raise the overall level to maximum.

Option 3:
Lower the frequency with higher ranges, to obtain a linear frequency response. Then raise the overall level to maximum.

To reduce the total power of the very deep frequencies, I could live with an additional applied high pass filter, but I definitely want to preserve the deep bass.

Please excuse my bad English and thanks in advance! ;-)
« Last Edit: May 08, 2018, 03:38:50 PM by Pittylabelle »
Current Gear:
Microphone: Soundman OKM Rock "Special Variant" (binaural)
Recorder: Olympus LS-12

Offline Ben Turnbull

  • Don't ask me about Audacity...
  • Trade Count: (15)
  • Needs to get out more...
  • *****
  • Posts: 8641
  • Gender: Male
  • A Clark Nova ate my lunch... Don't ask, just pee
Re: Post-Processing - Corrective EQ on AUD-Recordings
« Reply #1 on: May 08, 2018, 07:30:11 PM »
I don't know your mics.
I think you might be over thinking this.
EQ in baby steps.
Trust your ears.

Good Luck
Who came first, Dave King or Zack Galifianakis?...
-My name is Ben, and I'm an AQhaulic ranting hysterically in my dotage
Official Archivist for Jazz Central Studios, MinniMN
DPA4021/SKM140/NakCM1k/ADK A51s/Countryman B3/ATes943/SPc4/och... > TinyBox OT> PCM-M10/DR-680/R4/T+mod UA5 needs attention

Offline Jammin72

  • Trade Count: (0)
  • Taperssection Member
  • ***
  • Posts: 827
  • Gender: Male
Re: Post-Processing - Corrective EQ on AUD-Recordings
« Reply #2 on: May 09, 2018, 01:23:53 PM »
Solid rule in EQ.  Cut, cut, cut.  Then give it gas.  Works on live sound mixing and in post production.  Especially with 24 bit recordings.  We have a good amount of headroom to play with  You also don't need to shoot for totally linear as that's not how sounds actually appear "in nature".  Outside of problem frequency cuts or notching a small overall presence rise isn't a bad call for far field recordings, some mics actually do this on their own, but it's a personal preference call rather than an actual recommendation.

If you have the latest Izotope suites try letting the "Master Assistant" have a go at your files, I'm pretty surprised by what it comes up with on it's own sometimes after analysis.

And as noted above.  There's not a "correct" goal here, you're tweaking things to sound the way you would like to hear your own recordings.
Yes, but what do you HEAR?

Offline nulldogmas

  • Trade Count: (3)
  • Taperssection Member
  • ***
  • Posts: 766
Re: Post-Processing - Corrective EQ on AUD-Recordings
« Reply #3 on: May 09, 2018, 02:04:26 PM »
When I EQ, I literally close my eyes while raising and lowering each EQ band until it sounds optimal to me, then smooth out the curve some when I'm done. I don't know that it's the best method, but it doesn't seem to be any worse than any other that I've tried.

Offline Pittylabelle

  • Trade Count: (0)
  • Taperssection Newbie
  • *
  • Posts: 31
  • Gender: Male
Re: Post-Processing - Corrective EQ on AUD-Recordings
« Reply #4 on: May 09, 2018, 02:38:21 PM »
Thanks for your quick answers!

I thought a little graphic visualization through EQ operations can not hurt. ;-)

Can you also tell me something about my three workflow options?

Option 1:
Raise the frequency areas with lower ranges. As a result, clipping occurs.

Option 2:
Reduce the overall level first, to get enough headroom so that no clipping can occur.
Then raise the frequency areas with lower ranges. Then raise the overall level to maximum.

Option 3:
Lower the frequency with higher ranges, to obtain a linear frequency response.
Then raise the overall level to maximum.

---------------

OKM Rock Technical specifications:

Directivity: omni directional
Frequency range: 2o - 2o ooo Hz +/- 3dB
Sensitivity: 5 mV / Pa +/- 3 dB
-46 dB ref 1V / Pa +/- 3dB

Sound field: reversed
Maximum sound pressure (SPL): 142 dB
(142 dB, K = 3% U gr. As 7.5 V, R = 18 kOhm)

Equivalent signal / noise behavior:
dB (A-weighted), effective 26 dB
Peek: (CCIR 468-2 filter: 39 dB

---------------
Current Gear:
Microphone: Soundman OKM Rock "Special Variant" (binaural)
Recorder: Olympus LS-12

Offline Sloan Simpson

  • Trade Count: (1)
  • Needs to get out more...
  • *****
  • Posts: 3510
  • Gender: Male
    • Southern Shelter
Re: Post-Processing - Corrective EQ on AUD-Recordings
« Reply #5 on: May 09, 2018, 03:01:02 PM »
Thanks for your quick answers!

I thought a little graphic visualization through EQ operations can not hurt. ;-)

Can you also tell me something about my three workflow options?

Option 1:
Raise the frequency areas with lower ranges. As a result, clipping occurs.

Option 2:
Reduce the overall level first, to get enough headroom so that no clipping can occur.
Then raise the frequency areas with lower ranges. Then raise the overall level to maximum.

Option 3:
Lower the frequency with higher ranges, to obtain a linear frequency response.
Then raise the overall level to maximum.

---------------

OKM Rock Technical specifications:

Directivity: omni directional
Frequency range: 2o - 2o ooo Hz +/- 3dB
Sensitivity: 5 mV / Pa +/- 3 dB
-46 dB ref 1V / Pa +/- 3dB

Sound field: reversed
Maximum sound pressure (SPL): 142 dB
(142 dB, K = 3% U gr. As 7.5 V, R = 18 kOhm)

Equivalent signal / noise behavior:
dB (A-weighted), effective 26 dB
Peek: (CCIR 468-2 filter: 39 dB

---------------

Option 2 (or record at a lower level in the future)
Neumann KM-184> Tascam DR-680

Offline Gutbucket

  • record > listen > revise technique
  • Trade Count: (13)
  • Needs to get out more...
  • *****
  • Posts: 12302
  • Gender: Male
Re: Post-Processing - Corrective EQ on AUD-Recordings
« Reply #6 on: May 09, 2018, 03:26:45 PM »
Any of your options should work.  The issue is correct overall gain through the EQ.

If you get clipping after the EQ, you just need to lower the gain out of the EQ until all peaks fit within full-scale.  There should be a output gain or make-up gain control to do this, and that control is typically located on the right side of the graphic window.  It raises or lowers the output level of the EQ equally across all frequencies, essentially shifting your entire curve up or down.

If you get clipping in the EQ, you need to lower the gain into the EQ until all peaks fit within full-scale.  There might be a input level or control to do this, and that control is typically located on the left side of the graphic window.  If not, either reduce the level of the signal prior to the EQ, or just cut instead of boosting.

When I EQ, I literally close my eyes while raising and lowering each EQ band until it sounds optimal to me, then smooth out the curve some when I'm done. I don't know that it's the best method, but it doesn't seem to be any worse than any other that I've tried.
^
This is good advice.  Work on smoothing whatever shape you arrive at once you've dialed in whatever sounds appropriate, as long as doing so still sounds right after the smoothing. If you do the smoothing correctly, the result will sound about same or better. There is no sense in smoothing if it makes whatever you arrived at sound not as good!  The goal is good sounding audio, not nice looking curves.

Don't worry about only cutting. Boost or cut as needed to make it sound right.  I boost all the time.  Cutting only is more of a live thing, where high-resonance feedback can be a problem.  If a simple boost of the presence region is what you want, do that.

The main thing is keeping good control of signal level throughout the chain.
volition > vibrations > voltages > numeric values | numeric values > voltages > vibrations> virtual teleportation time-machine experience
made easy- >>Improved PAS table<< | made excellent- >>click here to download the Oddball Microphone Technique illustrated PDF booklet<<

Offline Pittylabelle

  • Trade Count: (0)
  • Taperssection Newbie
  • *
  • Posts: 31
  • Gender: Male
Re: Post-Processing - Corrective EQ on AUD-Recordings
« Reply #7 on: May 09, 2018, 04:54:40 PM »
Any of your options should work.  The issue is correct overall gain through the EQ.

If you get clipping after the EQ, you just need to lower the gain out of the EQ until all peaks fit within full-scale.  There should be a output gain or make-up gain control to do this, and that control is typically located on the right side of the graphic window.  It raises or lowers the output level of the EQ equally across all frequencies, essentially shifting your entire curve up or down.

If you get clipping in the EQ, you need to lower the gain into the EQ until all peaks fit within full-scale.  There might be a input level or control to do this, and that control is typically located on the left side of the graphic window.  If not, either reduce the level of the signal prior to the EQ, or just cut instead of boosting.

If I lower the gain into the EQ, I can not successful balance the frequency response until all peaks fit within full-scale. That's just the problem.

Here is a screenshot of the main window of iZotope's Corrective EQ:
https://drive.google.com/file/d/1MtlapOuzGBokFiaGV0iBMTi6oqr6DmjI/view?usp=sharing

iZotope shows the EQ very nice visually - so very user-friendly.

In the default setting, the (white) line is of course zero. This then starts directly at the RAW recording. Based on that, I can lift or lower certain frequencies. Therefore, the screenshot does not show the frequency response, but the increase or decrease of certain frequency ranges.

Example: The yellow bell is currently active:
Freq: 550 Hz / Gain 7.5 dB / Q (Bandwidth): 0.6

In my example, when I try to balance (raise) the valleys in the frequency response, clipping occurs. That's why one of my considerations was Option 2: Reduce the overall level first to get enough headroom.

Further questions:
But what happens if I lower the overall level? Do I lose Dynamics or something else in the first place?
What do you mean with "cutting"? Maybe a hard limiter?
« Last Edit: May 09, 2018, 05:16:32 PM by Pittylabelle »
Current Gear:
Microphone: Soundman OKM Rock "Special Variant" (binaural)
Recorder: Olympus LS-12

Offline Ronmac

  • Trade Count: (0)
  • Taperssection Regular
  • **
  • Posts: 157
  • Gender: Male
Re: Post-Processing - Corrective EQ on AUD-Recordings
« Reply #8 on: May 09, 2018, 05:13:22 PM »
A golden rule of EQ is to "cut narrow and boost wide", meaning to make cuts to offending frequencies as narrowly (higher Q settings) as possible, and make gentle boosts on a wider band (lower Q settings) of frequencies.




Offline opsopcopolis

  • Trade Count: (3)
  • Taperssection All-Star
  • ****
  • Posts: 1443
Re: Post-Processing - Corrective EQ on AUD-Recordings
« Reply #9 on: May 10, 2018, 10:35:49 AM »
Any of your options should work.  The issue is correct overall gain through the EQ.

If you get clipping after the EQ, you just need to lower the gain out of the EQ until all peaks fit within full-scale.  There should be a output gain or make-up gain control to do this, and that control is typically located on the right side of the graphic window.  It raises or lowers the output level of the EQ equally across all frequencies, essentially shifting your entire curve up or down.

If you get clipping in the EQ, you need to lower the gain into the EQ until all peaks fit within full-scale.  There might be a input level or control to do this, and that control is typically located on the left side of the graphic window.  If not, either reduce the level of the signal prior to the EQ, or just cut instead of boosting.

If I lower the gain into the EQ, I can not successful balance the frequency response until all peaks fit within full-scale. That's just the problem.

Further questions:
But what happens if I lower the overall level? Do I lose Dynamics or something else in the first place?
What do you mean with "cutting"? Maybe a hard limiter?

Nothing happens. It’s just quieter. I agree with gutbucket. Lower the gain of the track pre-eq, do your eq and then raise it to be the desired level afterwards. When they say cutting, they mean lowering frequencies with eq as opposed to boosting
Mics: Berliner CM-33, CA-14 card, CA-11 card & omni, AT-853, Sony ECM-907
Recorders: Tascam DR-60D, Tascam DR-05, Sony Hi-MD

Offline Gutbucket

  • record > listen > revise technique
  • Trade Count: (13)
  • Needs to get out more...
  • *****
  • Posts: 12302
  • Gender: Male
Re: Post-Processing - Corrective EQ on AUD-Recordings
« Reply #10 on: May 10, 2018, 10:55:29 AM »
Yes. Thanks.

Compare the raw and EQ'd versions at the same listening levels.  If you don't notice a higher noise-floor of the recording after the EQing and level adjustments, no problem.
volition > vibrations > voltages > numeric values | numeric values > voltages > vibrations> virtual teleportation time-machine experience
made easy- >>Improved PAS table<< | made excellent- >>click here to download the Oddball Microphone Technique illustrated PDF booklet<<

 

RSS | Mobile
Page created in 0.21 seconds with 33 queries.
© 2002-2018 Taperssection.com
Powered by SMF