@Christoph-Hart Thanks for the detailed info, I really appreciate it. I will start as soon as I get the time.

@Christoph-Hart said in The source code has different commit hash...:

Actually with the new AI superpower of just telling Claude to cherrypick a commit into another branch we might actually try to use the branch concepts as they are intended.

For those completely new to this, where can we find information on the steps to follow for setup and operation for Agentinc Coding with HISE?

@alexey_inlay congrats! Which server is this system using in the background? AWS elastic?

By the way, attachToProcessingSpecs broadcaster also seems to get buffer size!

@ustk This is it, I will try this thanks!

@David-Healey There doesn't seem to be anything related to sample rate in Transport Handler :(

Is there a function that can automatically run when the DAW's sampling rate changes? Or how can we achieve this?

@pratitghosh said in NAM Model switching logic:

@resonant Also what's Aida x?

It’s a cpu friendly alternative to nam. Not sure if it is more accurate than nam but it is pretty good. It works in Hise too.

https://mod.audio/aida-x-colab-training-guide/

@pratitghosh yes documentation will be helpful.

Are you using Nam or Aida x? Nam is very cpu extensive in stereo processing. How do you prevent this?
Also can you embed models as json scripts intı the plugin, or external files?

@pratitghosh I think you can use timer node to make a gain node to mute then back to 0 dB to make a smoother transition. This time there will be brief silence every time the knob is tweaked.

@pratitghosh you can just use bypass nodes for each model. And these modes will be connected to a xfader node, so it will switch between models

Edit: 100 models are pretty huge, maybe multiple xfader nodes, each one will switch 8 models

@ccbl Thanks for the tip. So like @David-Healey says, it is impossible to convert files each other.

@iamlamprey I've already tried this, the project freezes while loading.

NAM files cannot be embedded, which is a major problem. Hopefully, the neural node issues will be resolved as soon as possible.

@David-Healey said in Guide for Training on the RTNeural:

@resonant said in Guide for Training on the RTNeural:

Is it possible to convert nam models to Aida-x models?

ChatGPT says no

Google says yes :)

https://github.com/baptistejamin/nam-to-rtneural

But I think RTNeural must be installed first:
https://github.com/jatinchowdhury18/RTNeural

@griffinboy Thanks for the info.

Another one, Is it possible to convert nam models to Aida-x models?

Has anyone installed RTNeural on their computer and used it for training? I haven't been able to find a beginner's guide for it.

@iamlamprey Have you solved this?

@HISEnberg I recompiled the dll so many times. Also deleted the HardcodedFX and loaded again. No luck.

@Christoph-Hart said in Neural node bug on HardocdedFX as embedded:

@resonant fixed now. The problem was that it tried to load the network in the hardcoded FX before the first script compilation so it couldn't find it and there was an uncatched exception.

Now it prints this order of events to the console, so when you load your project you'll see something like:

Main Interface: Can't find network with hash 1234
Main Interface: Initialised network with hash 1234 (NetworkId)

So if you read that everything's fine.

A friendly bump. Same issue persists.
I believe same thing happens on the Global Cables node too.