Machine learning for signal processing is no joke. It's an order of magnitude improvement. The approach is different, too. Instead of approaching the signal mathematically with Fourier transforms and such, you approach it as part of a pipeline where you extract high level components such as pitch and timber. The ML model learns how to do that.

This field is going to change everything. Look what I've been able to do with some of the recent ML innovations: https://vo.codes

IMO, this one's newsworthy because whatever they did works in click-boom fashion- no messing with gain or knobs or fiddling- there's a checkbox for remove noise, and a slider for how much you want to remove.

I'd been trying to get rid of the noise in my mic forever- and RTX Voice just worked.

(That being said, I also had to patch the installer, but like who cares? Someone else did the hard part of figuring it out already, and at that point in time, it was 'beta' software, as well.)

AI just means deeplearning here. The past few years have seen great improvement when applying deep learning to a bunch of already established domains, such as noise reduction and background removal

It's newsworthy because it's actually good. Might be even exceptionally good if it's for average person with cheap microphone.

What other software would you suggest for good quality real time background removal? The ones I've tried I've been stuck with pretty bad artefacts.

The deep learning based noise reduction is of significantly better quality than previous algorithms.

Depends on where this product ends up, I suppose. If it learns and improves it's current performance with my admittedly noisy background, then the improvements on the technology would be pretty incredible.

I mean, isn't the whole point of the tech industry to iterate on old tech and make life better with each improvement?

And a Tesla is just a Power Wheels.