Deep Vintage - the AI-Powered Hardware Simulation Plugin Series by Three-Body Technology

[ThreeBodyTech]

This is a machine that endeavors to pass the human-ear Turing Test,
a deep-learning AI technology with a vintage heart.

Deep Vintage series is a suite of plugins that simulate real hardware. Utilizing the power of AI, every Deep Vintage plugin authentically captures the true 'soul' of the vintage gear.

Powered by Three-Body Tech's self-developed APNN (Audio Processing Neural Network), a machine learning technology specialized in simulating analog effect processors, Deep Vintage guarantees you the closest listening experience ever to the legendary hardware models, while ensuring a complete real-time processing.



How is it different?

We have been exposed to numerous emulation technologies: physical modeling, convolution... among many others. Regardless of what technics to engage, our ultimate goal is consistent: to achieve the highest possible fidelity in reproducing the sound of the real-world devices.

APNN was meticulously designed and trained to excel in this task, as its learning is based on audio signals from the real hardware units. Deep Vintage plugins are not merely 'theoretically correct modeling of circuits' nor solely based on harmonic structures or impulse responses; they capture 100% 3D feels in the analog sound.

How does it work?



To put it simply, APNN is a neural network specifically optimized for audio processing, which is composed of a massive combination of audio processing modules (EQ, compressor, overdrive, etc.).

To capture the essence of a hardware model, APNN will automatically adjust its structure and parameters until the difference between its output and the hardware's output progressively diminishes.

Ultimately, APNN achieves an error signal of phase cancellation at about -40dB to -75dB (on the training set, not on the validation set). With this exceptional level of error control, which surpasses even the variance between different production batches of the same hardware model, we can confidently say that APNN is capable of 'deceiving' human ears.



Nothing in your DAW comes closer to the real hardware than the APNN-powered plugins.

Proudly introducing our first lineup in the series, including 6 plugins inspired by legendary models:
Brit 73 AI
US Rack AI
Thick Pre AI
Green AD AI
Tape S AI
Tape M AI

Hear demos & start your FREE TRIAL at HERE.

INTRO OFFERS for limited time.
Now get 3-in-1 Preamp Bundle or Tape Bundle with 50% OFF!
SAVE MORE on 6-in-1 Complete Bundle!!

[jamcat]

Brit 73 AI = Neve 1073
US Rack AI = API 550B
Tape S AI = Studer A800

What are the others?
Revox or Ampex?
Chandler Ltd?

[bungle]

Wow, that is a whole lot of word salad on that site and OP!

[qtheerearranger]

that is the germanium preamp from chandler i believe. just using presence as a phrase instead of feedback. could be wrong. probably am wrong.

[Igelhaft]

The last two are some Burl and what is speculated to be a Telefunken M15 or M20.

A lot of people on GS seem to be excited about these. I haven't tried the demos, but based on the (excellent, full credit there) comparisons provided on the site I just don't think these are there yet.

Maybe when I eventually give them a shot they'll surprise me but they sound like... plugins, imo. Not a bad starting place though, if they continue to improve.

[jamcat]

that is the germanium preamp from chandler i believe. just using presence as a phrase instead of feedback. could be wrong. probably am wrong.

I don’t think you’re wrong.
It’s what I suspected as well.

[bmanic]

Very brave to have the A/B comparisons. Unfortunately they show quite clearly how different they are. This tech is not quite there yet. Having said that, I actually preferred some of the AI versions in terms of sound aesthetic on some of the clips. So who cares?

[imrae]

Ultimately, APNN achieves an error signal of phase cancellation at about -40dB to -75dB (on the training set, not on the validation set). With this exceptional level of error control, which surpasses even the variance between different production batches of the same hardware model, we can confidently say that APNN is capable of 'deceiving' human ears.

Hmm, the validity of this claim really depends on what the diff sounds like. Different production batches would give slightly different filter corner frequencies, or different asymmetry in saturation, or different program-dependent ballistics. If the model adds -60dB of quantisation noise, or a whistling signal, or extra band-limiting... that's not equally likely to go unnoticed!

But if you have got the "artifact-like" errors to a lower level then this sounds very promising! Will be interesting to trial.

[CapnLockheed]

Hrmn....looking at their VSTi's one might assume this company is based in China.
I hope their code is not being crunched by AI child slave bots.

[gentleclockdivider]

You know what really bothers the shit out of me , it's these AI narrated voice overs.
If you want to expose your product at least take the time to do an actual voice over .
It's so damn cheap