Izotope Rx Remove Distortion
Muredac Ford
I've been asked to try to save a few days of recording (interview with boom and 1 radio mic): the radio mic is always (four days of recording, I couldn't believe it!!) saturated/distorted; to me, it seems like if the production mixer/boom operator had set up the receiver to line level but connected it to a microphone level input in the recorder.
If I cannot save it with my tools, I advised the producer to go in another studio that have the big Cedar machine (the Cedar Cambrige): anybody has any experience with the cedar cambrige (I used only the DNS) that's willing to share and know if it could save those saturated recording?
Most Declipper plugs won't work in this case, because they're designed to work with the very straight, minimal harmonics of digital clipping. In this case you've got analog distortion which isn't quite the same. One thing you CAN try is using a De-Crackler plug-in. I've found the Waves X-Crackle the best for this, but the WaveArts works as well, and presumably the RX one (I haven't had an opportunity to try it yet). Try setting the values to maximum and see what happens. Follow that with some negative compression (i.e 1:2, 1:4 ratios )- I like the Waves R-Comp for this- and you can often minimize it, albeit not remove it...You can also try doing the negative compression first, sometimes that serves better.
De-clipper tools make an educated guess as to what the peaks should have looked like, and rebuild them based on what audio information is available i.e. the waveform just prior to distortion. Apparently they are only really effective when there is light clipping/saturation, and you already mentioned that you have used Izotope RX so I'm guessing that the saturation is pretty bad.
I think CEDAR is your best bet without re-recording the whole thing. Never used it myself but apparently a lot of the noise removal features are modular, or in other words they are add-ons to the basic package. If you do go down this route make sure the studio you use has the de-clip module. From what I have heard/read CEDAR is in a league of it's own, so if that doesn't fix it then nothing will.
Used in combination with another successful approach, you might help the quality at least a little with some eq, because whatever frequencies are emphasized by the mic and pre-amp, when gained up too far, tend to splatter all over and really dominate when squashed. You may also find that the crackle of the distortion lives predominantly in one frequency range and you can knock it down a little with eq. (Or try the C4 in this capacity.)
Desperation move: try blending a little of the lav mic in with the boom for clarity. You can experiment... maybe band-pass the lav's signal severely so you're left with mostly consonant sounds and sibilance, and see if you can blend this in to help make the boom audio more intelligible in the quieter spots.
You can try to fix it by editing the volume automation. Where you feel the signal is too much compressed, reduce the level. It is a painful process. I once made it for a 100 min documetary Maybe you can use RX as a guide. You schould see gain manipulations outside the human voices spectrum in the spectrogram.
I wanted to start a topic that focuses on the parameters of these two upsamplers in A+ and perhaps even get some objective measurements of Fourier transforms (impulse response graphs) from those people who can capture them. My experience comes from these software upsamplers as they are implemented in Audirvana Plus 3.2 - But I don't think there's anything keeping others with different players from chiming in here.
To kick things off, I'm using @copy_of_a 's extreme settings for iZotope as shown below. In order to avoid aliasing with his settings, you have to use a software plugin to further lowpass the signal. In my case, I'm using FabFilter ProQ2
For SoX, I've chosen the below based solely by ear. No LP filter is needed with an EQ plugin since the filter is much steeper it seems. Admittedly I haven't put near a much time into SoX as I have with iZotope:
Overall, the biggest difference between these is the inability to finely tune the filter steepness in SoX. I've read some anecdotal things and tend to agree with most of them. For instance, Sox seems to present a great image/stage/separation, but comes at the expense of some more ringing that can be heard in the "shimmering" highs on cymbals. What's interesting though, is that the preringing doesn't seem too bad when I scrutinize the timing of a kick drum - hearing first its batter head which should be followed by its resonate sub bass thump. I find that my setting in iZotope still perform best here - but the SoX isn't nearly as bad in the lows as with the high frequency "smearing/shimmering" I hear. But this could just be my settings, more time will tell which is the better upsampler for me.
without resampling you are of course seeing the unaltered transition-/stopband of the original file (which makes sense since the resampler is bypassed when you select "none" in the upsampling section of A+'s prefs).
I noticed similar aliasing and should've taken a screen shot of it. It's weird as the aliasing pulses up to -45db but quickly falls. This even happens with a bandwidth setting of 90. But using fab filter, I can presumably remove some of that aliasing with a LP filter at 96db/octave. Doing this, I still prefer your "extreme" izotope settings to any I could produce in Sox.
Sox is initially appealing because it sounds "euphonic". Perhaps this pulsing aliasing is causing intermodulation distortion that sounds good to some people? Still I'm not sure I can hear the aliasing. I'm more concerned that it could be damaging my electronics. Is there any validity in this concern? Also, what does aliasing sound like to you? What cues should I be listening for?
In terms of development and maintenance, does anyone have an opinion on which of these upsampling software products is leading the pack here? SoX is open source, but that doesn't mean it has any less potential to go up against the "pro audio" products.
@matthias, you were right. I was able to observe the rolloff steepness of the SoX filters at various settings using white noise and zooming in on FabFilter's spectrum analysis between 10 and 22khz. At the highest setting of 99.5 it is a "brick wall", whereas at the lowest setting of 74, there is a definite rolloff. What's interesting is that even at 74, the rolloff isn't very severe. It kicks in at 18K, much higher than I'd expect since 74% of nyquist at 44.1 is 16K approx. This is also what I'm hearing, zero loss of high frequency air, and I have verified up to 18K hearing by an audiologist (kinda freaky for my 38 year old hearing system).
Also, the ringing doesn't seem to be reduced by very much even at the 74 setting. Sometimes I think I like SoX, other times I'm convinced that I'm getting the better staging and separation of instruments at the expense of a sloppier kick drum attack. I guess we can't have our cake and eat it too, huh?
Initially I thought I preferred the new SoX setting. Now , not so sure ? SoX can sound slightly artificial and euphonic at times to my ears . Does seem quieter with more clarity though when listening to my Classical and Jazz piano recordings .
Some of these negatives are somewhat mitigated though when I use SoX without oversampling . With oversampling activated I find the treble through my Invicta DAC can sound somewhat thin and glassy or strident with a loss of mid- range tonality due to an increase in euphonics .
I've given SoX some more time and I like its staging (width) and liveliness, for lack of a better term. There's more shimmer in the highs and it's not as dry as iZotope. I agree with @Noodle that SoX can sound euphonic, reminds me of the "exciters" on DAWs that add some sparkle by increasing the overtones/harmonics of cymbals. It seems to spread a bigger canvas over my head (I'm a headphone listener).
yes, according to the Audirvana manual, this is to prevent the DAC from clipping. But my DAC has analogue filters that activate when DSD is sent to it. It has 3 positions: 0,-3,-6 db approx. I have it set to -3 currently.
I'm a long time A+ user and have tried many different filter settings over the years. I gave your iZotope setting a try and must say that it sounds fantastic. I'm exclusively listen to loudspeakers and have owned Magnepans for over a decade a now currently have a pair of 3.7i's with a REL sub. I state this as so many now are exclusive headphone listeners.
I saw your screen shot you posted regarding your SOX settings and noticed that had a DSD DAC. Do you feel that the SOX filter should primarily used with DSD? It seems that the SOX filter lacks just a touch of warmth. When I did audition a DSD DAC a few months ago, my impressions were that it was a touch on the warm side of things so I've come to the conclusion that SOX is a better fit for DSD. Do you agree with this assessment?
On a different note, I noticed the I can no longer adjust filter setting by directly inputting settings into the box to the right of the sliders in A+ but must us the slider to make changes. Have you noticed this as well?
I think that's a bit too much of a generalisation. The Sox and iZotope resamplers both have a multitude of settings. Most of what one can do, the other can too. If they sound different even with the same settings, that would be interesting.
Everything being equal, you really should not need any sort of repair system when recording voice-overs professionally. The studio environment should be perfect, the recording equipment should be top-line, the sound engineer should be highly experienced, and the talent should have the perfect voice.
A General Guide
Recording Dirt
Recording My Poetry
Preparing Your Manuscript
Plan your audiobook as you write
Using iZotope RX 6 with Cubase
Editing in Cubase demonstration
A bit on Punch & Roll
Should you delete breaths?
Setting up your Studio
External Articles & Resources
Punch and Roll