Processing Audio for Visualization

[hankTremain]

I'm looking for some help processing audio to optimize it for visualization and this forum seems like the place I may find people who can point me in the right direction.

What I'm looking to do is to process any sound-wave as to outputs a waveform that behaves like a 'VU meter'. So, in other words, has the negative signal inverted and summed with the positive, and a gradual fall off from the peaks. I've attached an jpg illustration to demonstrate what I mean.

audioProcessArtboard 1.png

This may seem like a strange thing to do to a sound-wave, as the result will be unlistenable. However, having the audio processed in this way will allow me to input it into video creation software like 'Blender', use it to drive time based animation, and have it run sample for sample in time with the audio of the video.

All DAWs must use math to do something similar to this for the signal monitor meters on all audio channels. My question is, is it possible to output this processed signal as an audio file (.wav etc)?

Anyone got any ideas?

[imrae]

.wav files aren't especially complex, they can certainly hold slow DC-coupled envelopes like this if that's what you need.

I would point out that this kind of "envelope" processing is very common as part of the design of dynamics processors such as compressors, expanders and gates. You may want to look into how those are typically implemented.

[hankTremain]

Thanks so much for the reply - that's very encouraging.

Regrettably I have next to no developer knowledge - so building a script to process this type of change to an audio file is out of my scope without help.

I wonder how far I might get using pre-existing plug-ins within a DAW?

Using an extremely fast delay and tuning the feedback might help me achieve this averaging between peaks with a slow fall-off. However, if this is done before inverting the negative parts of the waveform the phase relationship between delays would likely cancel itself out - no?

If I could asymmetrically clip the waveform to 0db for any signal below that point - I could duplicate the audio to a separate track, invert the phase of the duplicate, apply the clipping to both tracks and sum them together to achieve this first part of the process. However, I don't know how to do this within a DAW - I feel like a "brick wall" limiter could do the trick if only I could find one that was asymmetrical (and didn't effect both sides of the waveform). Maybe a DC offset to push the wave completely into a positive signal? - though again I don't know how to do this within a DAW??

Thinking about it, that seems to be the way?
Maybe, if I were start by summing the waveform with a positive DC signal, add a sample fast delay, and then cancel out the DC offset; I'd have something close to the desired result?

[BertKoor]

Go go Duck Duck "bIender audio visualizer envelope follower"

Result #1: https://lesterbanks.com/2020/02/create- ... h-blender/

More info with those keywords on YouTube...

[hankTremain]

like this:

audioProcess_2Artboard 1 copy.png

[hankTremain]

AUTO-ADMIN: Non-MP3, WAV, OGG, SoundCloud, YouTube, Vimeo, Twitter and Facebook links in this post have been protected automatically. Once the member reaches 5 posts the links will function as normal.

Go go Duck Duck "bIender audio visualizer envelope follower"

Result #1: https://lesterbanks.com/2020/02/create- ... h-blender/ (https://lesterbanks.com/2020/02/create-an-audio-visualizer-with-blender/)

More info with those keywords on YouTube...

Thanks for the reply.
I'm not at all a noob when it comes to Blender, and am not looking for a tutorial. The "Bake audio as F-curves" operation that CGMatter uses in this video is exactly the feature in Blender I am looking to exploit. However, there are a number of issues with this if you use raw audio that I am looking to avoid by processing the audio in the manor I'm proposing. My issue is how to achieve such a process without building my own software to do so.

[BertKoor]

Audio levels are often expressed as RMS: Root of Main Square. Look it up how to calculate.

Since the target is video, a frame rate between 25 an 60 per second seems reasonable.
What if you simply take the max of the signal within such a block?
Me thinks that's good enough for Rock'nRoll.

[Tj Shredder]

Its dead simple. Its called envelope follower and is used all over the place as mentioned inside compressors and alike, but also solo for side chaining purposes. Which DAW are you using?

[bk]

Audio levels are often expressed as RMS: Root of Main Square. Look it up how to calculate.

Nope. It's Root Mean Square.
Not trying to be a dick, but if they are to look it up, it helps to know exactly what to look up.

[BertKoor]

Thx bk for the correction.

[bk]

Thx bk for the correction.