Hi folks
It is possible to make like de-filtering job in SL10 ?
What kind of filtering would you like to undo ?
Hi Robin
In most usual sense, like removing effect from standard LP filtering, or can say - increasing (or maybe even synthesizing ? lol) high spectra for “naturally” filtered material like old\tape recordings. Respectively wondering about auto-processing, in some adaptive way, not just as manual EQ’ing.
You could have a look at EQ Match: EQ Match
1 Like
Ok thanks, will dig into this feature.
Do you mean something like this?
Quite an old plugin now, but would be great to see this functionality in SL (if only we knew how they do their magic!)
Yes i meant exactly that (just did not want to mention lol).
What regards his (Zyn Unfilter) magic, once time i trialed it intensively, and had interesting observation, that i can achieve the same high spectra’ unmasking thru drastically EQing.
Without belittling its merits, since I am sure it saves time, and it also has the function of using IR profiles, which I think is very convenient. But so far, it seems to me, there is not as much magic there as it might seem at first glance.
And I assumed that SL should also be capable of this.
In Zynaptiq Unfilter the unmasking effect is the result of a process that is dynamic over time (and possibly in the frequency domain?).
That is not possible to achieve with simple eq-ing or eq-matching based on static spectral ”snapshots” of a given time/frequency frame.
Hmm ok. Maybe you right. The only thing what i meant, i had able to reproduce the same “unmasking” ability which was shown in this demo video, with just usual EQs.
https://www.youtube.com/watch?v=7thIB9VX35k
The material on which I tested was also like a piece of a song, which I strongly suppressed with an LP filter. And result, as I already said, was similar. Perhaps I missed some details, I won’t argue.
But in general, that test was enough to cool down a little for purchasing. (Although, I still wouldn’t refuse it, but not at regular price.)
UPDATE: But yes, resp can assume that if we are dealing with material that is not statically covered with filter along its entire length, but only in places, then just EQ will not do the same.
(But for me, this effect of removing the filter itself was most important then).
Yeah, I have Zynaptiq Unfilter and it is a cool app. To achieve the best results with it you should have the ”learn” function active, for letting the app continuously adapt the eq-ing to the variations in dynamics and frequency ranges along with the song.
If you turn off the ”learn” function it will act more like a (non dynamic) regular eq or perhaps an eq-match app (like in SpectraLayers).
It will of course be possible to unmask certain frequencies in that mode if you find the right frequency to adjust with the eq curve, but that might take quite a while to identify. And it will probably change along with the song with respect to the actual dynamics/power and frequency contents.
1 Like
What I would like to see is a DeFilter that generates content based on the contents in the signal. So instead of something that’s just “dull” where you can add an EQ and boost the highs I’m talking about a bandwidth limited mp3 or phone recording where it’s basically brickwalled above a frequency and the plugin then calculates what the overtones would be above that cutoff. I think this is at least partially what Zynamptiq’s plugin does, and iZotope RX too has a module for it.
Additionally and importantly that should also be possible for the low end.
Imagine getting a bandwidth limited phone conversation or dialog that is limited both high and low and then regenerating that content… Or, getting a piece of vocal/dialog where there is a hum and high end noise at the same time and after you’re done denoising it you’re lacking both lows and highs - again regenerating both.
I can’t remember exactly how but I’ve actually done this before using Spectralayers. A lot of the old content on youtube is badly compressed and I figured out a formula to (sort-of-speak) remaster old audio using Spectralayers (including reconstructing additional harmonic content)
You can kinda already do this but to ARTICULATE and explain it (in a coherent manner) is a different story.
I’m not a computer right now but I believe you can use the harmonic selections tool along with the combinations of transformations tool to move the selections up and down (from there you should be able to figure out what to do from that point). What would be a useful feature is the ability to generate(process) tones/partials based off numerous selections(and or multiple selections). That way it wouldn’t be so tedious to generate(process) a separate tone/partial for each individual selection.
1 Like
In general, it seems to me that in era when exists picture and even video generating (ofcourse, by using large data of prepared library), make similar thing for generate lack frequencies in sound, should be deal not more harder. Would be cool to see such in SL