top of page
Search
Steve

TONEZONE3000: Training Made Simple

NAM is the fastest, most accurate--and now (I'd argue), easiest--way to make a neural model of your gear.


TONEZONE3000 is a third-party website that puts the standard snapshot training process in a streamlined, easy-to-use website. Like the GUI and Colab trainers I've developed, users can reamp a standardized "sweep" signal through their gear, then upload the recording and follow the steps to get a model in minutes.


Accurate

TONEZONE3000 is built from the latest release of the standardized training code that I've open-sourced. You'll get the same high-quality NAMs from it as you've gotten from the other trainers. TONEZONE3000 also exposes a variety of "advanced options" that can help with tuning the training process if you'd like.


Fast

TONEZONE3000 uses nVIDIA A40 and GeForce RTX 4090 GPUs, which are far faster than the T4 GPUs available for free on Colab--or what most users will have a home, for that matter. It takes under 3 minutes to train a model for 100 epochs from start to finish.


Easy

One cool thing about TONEZONE3000 (and perhaps a first in the space of standardized trainers) is that you don't need to be able to reamp to make a model. If you can simultaneously record yourself playing your amp as well as the DI signal of your playing, you can upload that pair of files to the website and make a model based on your own custom dry/wet data--which I'm going to go through in the next part of this post.


Walkthrough

I personally excited because of how easy the whole process is. To demonstrate, I'm going to go through the steps that a beginner might do to make a model. Here's my setup:


My setup. My guitar goes to the DI box (blue cable), where it's split and fed to the amp and the interface (orange cable). The mic goes to the other interface input (black cable). REAPER is open on my laptop and is ready to record from both inputs.


On my laptop, I've got my DAW with two tracks set up: one for the DI, and one for the mic. I hit record and play for a couple minutes.



Next, I head over to https://www.tonezone3000.com/ and upload the recordings as a dry/wet pair



Clicking "Create Tone", the website walks you through the next steps. If you don't want to get into the details, there's very little you have to do:



Press "Create Tone", and training begins! You'll be taken to an account page where a progress bar shows how things are going:



Once it's done, you can listen to the result in the browser and download it if you're happy:



Then just pop it into a plugin with NAM snapshot model integration, and you're done!


If you want the model I made, you can try it here.


Enjoy!

1,473 views

Recent Posts

See All

Plug-in version 0.7.11 is released

I've released a new version of the NAM snapshot plug-in. You can download it on the users page . This release fixes a bug that users were...

NeuralAmpModelerCore v0.2.0 is released

There's a new version of the core DSP library for NAM: version 0.2.0 . This makes a few breaking changes but also cleans some things up...

Trainer version 0.10.0 is released

I've released a new version of the Python package for training models! As usual, the newest version is automatically used in the online...

NEURAL AMP MODELER

©2024 by Steven Atkinson

bottom of page