You have the full functionality of Thimeo Perfect Declipper in Thimeo Stereo Tool v. 9.40, in which you can use Configuration/User Interface/Operating mode/Expert to set 'Absolute highs' very precisely. (In 'Extreme Tweaker' mode you can set 4x oversampling to Absolute Highs).
Also check "Follow Delossifier' if you use Delossifier (which compensates for lost midrange due to psychoacoustic filtering of lossy files) along with Absolute Highs.
Delossifier does similar job to Apulsoft ApUnmask.

 

Last edited by a moderator: Nov 30, 2020

Okay understood. Thank you for clearing it up.

Here are the settings of the individual plugins I used:


And here's Acon Vitalize:


I'm not sure if this is the best combination but apparently, it's working.

 

Could you please expand on that ? How do I do that ?

 

It filters out additional quantization introduced by lossy codecs, which simplify the data by quantum rounding, in order to additionally shrink the file size.

Both optimized using DDMF Plugin Doctor.
Vitalizer setting depends on the bit rate of your lossy file. According to your spectral graph, your file is of somewhat lower bit rate (I'd say 128 or 160 kbps).

 

Last edited by a moderator: Nov 30, 2020

Yeah! there's a difference in the output. I like the end results. Thank you pratyahara, without you, I would have never gotten this far. I hope you become the president of US or whichever country you reside in.

@pratyahara will it be possible to get some presets for these two plugins if you have some ? Forgive me if it's wrong to ask and please ignore if you don't want to share.

And big thanks to all other members of the community who shared their expertise in the thread to get me going. Thank you guys, you are awesome.

 

You'll also need Zynaptiq UnFilter and Zynaptiq UnChirp to complete the job.
Here are all the presets you need.
https://www80.zippyshare.com/v/PSIDyKpm/file.html

For obligatory declipping use Terry West Relife with no 'Preclean', and 'Peaks' set at 1. 'Post Gain" set at +10. Make two passes in a row. Make sure your max. file peak is -6dB first (no more, no less). You might get gain change in the end, so you should adjust the gain to normal after this procedure for further processing.

 

Last edited by a moderator: Dec 1, 2020

Got it. This completes the project. I can never thank you enough for the help and the expertise. With your presets I can really do this like a Pro.

I wish you all the happiness in the world

 

I often disagree with you but this time I just take my hat off with your dedication to help a fellow forum member. Well done

 

Whoah! ApUnmask is applied to full-rate audio PRIOR to converting to lossy.
From website:
Suggested Workflow

• Insert apUnmask on the master track or an individual track to be enhanced.
• Adjust the master gain until a reasonable amount of delta signal is visible.
• Toggle the delta listen button to hear the "ghostly" sound what would be removed by the opus codec. This is what normally lives in the shadows of your music.
• Experiment with the resolution knob to find the most interesting components.
• Tweak the eq curve to extract and boost the good stuff.
• Turn off delta listen and enjoy the result.

Delossifier cleans up MP3 files!

 

Delossifier does two things: 1. kills pre-ringing 2. fills Spectral holes.
In my post I had in mind Spectral hole filler module:
From the author's site:
"Spectral hole filler panel
Controls the Spectral Hole Filler part of Delossifier.

MP3 and MPEG2 encoding at lower bitrates causes holes in the spectrum - typically, the louder sounds are encoded until there are no bits left, anything that's not encoded yet at that point is turned into silence..."

So the Spectral Hole Filler module of Delossifier does not 'clean up' MP3 files. It adds up a part of missing spectrum (this procedure can't be called "cleaning (up)").

So it is just an opposite way of using ApUnmask (a posteriori, vs a priori). The same procedure is valid if you prepare a lossless file for lossy encoding (to 'back up' part of the dynamic spectrum from pending spectral holes due to lossy encoding standard procedures) or you are trying to 'revert' an already lossy encoded file back to a lossless one.
Both ways of usage do not guarantee a 100% success because they are partly based on guesswork.

 

Agreed! Sorry, that's what I meant by "cleaning up"! "Repairing shit!" Whatever!

Disagreed!
Please read again what ApUnmask does. You really have misunderstood.

 

Forgive developers, they often do not know what their software is able to do.
I forgot to mention that declipping works better (as de-clicking does too) with file reversed.

 

Oh yes, totally agree!
"Able", I am not denying. Still very different to intended use as stated by the devs.
It's like an expander of the freqs that we know will get lost when it's eventually compressed to lossy.
So it is applied to the full-rate wave file.

 

Reverse is not very different, but just opposite in an opposite situation, so it is completely logical.

 

Applying it to fullrate audio, or applying to a lossy mp3, is not a "reverse" situation. Logical or otherwise.
The fact that using ApUnmask on a finished master will subtely emphasise quieter things prior to it becoming lossy, yet applying it to a lossy mp3 will not create the frequencies in the same way as a plugin that was designed for mp3s, should be enough to prove the point of why I'm just restating what the devs have said.

 

Devs say whatever they have in their one-way frame of mind.
Applying any software to a lossy mp3 will not create the frequencies to bring it back exactly to the original lossless file. But this one helps a lot, and currently there is nothing better to do the job. It does it much better than the dedicated module 'Spectral Hole' in Thimeo Stereo Tool.
It would not hurt anybody to try.

 

Hi @pratyahara , unfortunately the link you posted has expired. If at all possible, would you please be able to provide a fresh link for the presents you kindly shared as I'd be interested to learn from these, along with your other posts on similar topics to this? Thanks

 

I was doing a project for a client, it dragged on for several years. During that period I changed my computers at least 3 times (even changed my DAW's twice) and I lost some of my source files. But my client had all the MP3 previews I sent him in the process. So... when making mastering for the commercial CD, I used those MP3's for several songs. Of course I did some EQ, some exciting etc. etc., but anything above 16 kHz had been lost anyway. It could be seen clearly on any specytrum analyzer.

But my client was delighted with the resulting master. He praised it!

And the same person told me later, that he doesn't like to listen to the music from YouTube, because he can clearly hear it is of inferior quality and lacks highs.

 

If you tell someone what (specific flaw) to listen for in a recording, before you play it; they will almost always "hear it". Whatever that "It" may be. The mind believes, so the ears hear.

If you take into account that human hearing typically is only up to 20/21 kHz; you lost the top 1/4 of the recording, but then did your best to fix it with exciter, maybe some other processing also? It isn't really that surprising to me, that you were able to get it to "acceptable quality" to the person you did the work for.

 

I have NOT lost the top 1/4 of the recording. I lost the top 3-4%.

Humans perceive frequencies logarithmically. A young, healthy human can hear up to 10 octaves (~20 Hz to ~20 kHz). Losing everything above 16 kHz means losing less than half an octave (14 kHz would be half an octave). So I lost 4% of the recording at most.