Nu-Trix The Synth Guy Throws ReSing Into the Rave Bunker: AI Vocals Get Real

Nu-Trix The Synth Guy kicks things off by making it clear: ReSing isn’t your average cloud-snooping AI plugin. Everything happens right on your machine, so your embarrassing vocal takes aren’t going anywhere. That’s a proper win for privacy and session security—no random servers, no data leaks, just you and your CPU sweating it out together.

But there’s a catch: you’ll need a decent computer, preferably with an NVIDIA GPU if you want to train your own models without waiting for the next ice age. Nu-Trix runs it on a MacBook Pro M1 and survives, but warns that heavy-duty modeling can drag. Still, for most users, the local processing means you keep control and don’t have to worry about your creative gold ending up in some AI training set for the next pop robot.

ReSing isn’t just about making you sound like someone else—it’s about total vocal flexibility. Nu-Trix shows how you can swap your own voice for a model, create harmonies, or even build a whole backing choir from your dodgy iPhone takes. The plugin lets you pick from a line-up of human and cartoon voices, and even a few instruments, so you’re not stuck with the same old presets.

You can tweak everything from the range (auto-transposing to fit the model) to the energy and accent, and even blend between the original source’s phrasing and the model’s delivery. Want your backing vocals to sound like a chipmunk or a soul diva? No problem. The plugin’s controls let you dial in just how much of the model’s character you want, so you can stack harmonies or build a Frankenstein’s monster of vocal timbres.

Nu-Trix walks us through the nuts and bolts of using ReSing in Ableton Live 11, and it’s all about drag-and-drop. You can’t process in real time—this is strictly offline territory—so you’ll be bouncing files back and forth between the plugin and your DAW. He drops in his iPhone-recorded vocals, converts them to WAV, and gets to work previewing and processing with different models.

There’s a bit of a learning curve: drag the wrong file type and you might crash the plugin, so stick to WAVs if you want to avoid a toaster-fight with your DAW. Nu-Trix suggests prepping all your clips in advance and batch-processing them for maximum efficiency. The plugin spits out new files for every model you try, so you can quickly build up a library of alternate takes, harmonies, and wild experiments.

The workflow isn’t flawless—occasional glitches and mispronunciations pop up, especially if your source material isn’t clean. But for those who like to tinker, the process is fast enough to keep the creative flow going. And let’s be honest, watching Nu-Trix juggle files and voices is half the fun—some things you just have to see in the video to get the full chaos.

Nu-Trix wraps up by tackling the big questions: is this AI tool ethical, and do you actually own your creations? ReSing keeps everything local, pays voice model contributors, and ditches the subscription model for a one-time buy. That’s a rare combo in the AI world—no monthly bleed, and no mystery data harvesting.

Most importantly, the plugin keeps the artist in the loop. You still need a real performance to drive the AI, so it’s not just spitting out soulless clones. For producers who care about both creativity and ethics, ReSing lands in the sweet spot. If only more plugins played this fair.