Neural Amp Modeler (NAM) running on Raspberry Pi

Подписаться 272

Просмотров 19 тыс.

50% 1

This audio was recorded using the NAM machine-learning amp modeler running on my DIY Raspberry Pi 4 based guitar pedalboard.
Hardware used: Raspberry Pi 4. Hotone Jogg audio interface. Hotone Ampero midi controller. Wio Terminal (just for display).
The software is a custom Jack audio client - code is available here:
github.com/mikeoliphant/stompbox
I also have made an LV2 plugin available here:
github.com/mikeoliphant/neura...
There are a few more details on my blog:
blog.nostatic.org/2023/03/neu...

Развлечения

Опубликовано:

24 июл 2024

Ссылка:

Скачать:

Готовим ссылку...

Добавить в:

Мой плейлист

Посмотреть позже

Комментарии : 50

@mistergrimm6515 Год назад

That's great! Are you planning to upload a tutorial about this project in the near future? I believe It would be really appreciated!

@juliankirby9880 Год назад

Yes. Please upload a tutorial like this commentor requested. I grow plants. I am not a gifted computer user. I can only do stuff by following prompts or very explicit instructions. Not even joking, I am good at learning, bad at doing.

@rocknrollboise 6 месяцев назад

Sounds like they're just planning on keeping this all to themselves (unlike Mr. Atkinson who designed NAM for the masses). Sigh.

@stijnxd Год назад

Genius, I would veeery much appreciate a full tutorial on this!

@LeonTodd Год назад

Great work!

@steveparkes Год назад

Nice, after I've finished a couple of things around the house today I was going to try and build NAM for the PI. Nice to see somebody has already done the work :)

@jacobocameloherrera2781 Год назад

Genius. . . mate, simply genius ! neat idea !

@DomMcsweeney-DoomPlugs Год назад

This is a worlds first I do believe! Excellent indeed. =)

@ryanburchett8455 Год назад

Awesome idea

@forrymight274 Год назад

Fantastic!!!!!

@moroboshidan7960 Год назад

Yeah! That's what we need, Open Source modeler/multifx.

@jcsk8 Год назад

Looks promising. I love cheap good stuff.

@bendyboy Год назад

Now this sounds like project I would like to complete. Congrats on your work! Any chance of a how to? Ta

@dyonight Год назад

wow! I have so many questions!

@siddonfer Год назад

Its Great . Can you show step by step how you does ? tks

@piotrkarpienia4779 Год назад

We the people beg you for a tutorial. 🎉

@vgmoulin Год назад

Great! I've been looking to build something alike for a long time. The NAM was the missing link in my chain! Would you share your building process?? Great job!

@NostaticSoftware Год назад

Raspberry Pi 4. Hotone Jogg audio interface. Hotone Ampero midi controller. Wio Terminal (just for display).

@kennethsmith5383 Год назад

I have a Raspberry pi that I have been wanting do stuff like this with.

@giorgossynetos9723 Год назад

Shall we expect a tutorial on how to do it?

@BrainDeadTX Год назад

Man I wish I was this smart

@jonathanarnold2225 Год назад

Haha nice!

@adairo16 Год назад

This project will work with a touch hdmi display (amazon ones) in order to modify presets via touch screen?

@Sascha_Franck Год назад

Any words on latency? Like a measured value?

@KiLLJOY1056 2 месяца назад

Bro this is exact idea i had so i looked up Pi on Pedalboard and im so glad it works. Do you notice any latency at all?

@SoullessMindful 3 месяца назад

Do you have a recommendation for an audio interface that could be used instead of hot one, that has more I/o, low latency and works well on pi?

@rickard9294 Год назад

I have a question that I hope you can answer. What latency settings did you use and how approximately how much CPU power did it consume?

@NostaticSoftware Год назад

I'm running Jack audio with a 64-sample buffer. NAM "feather" models eat up most of the Raspberry Pi 4 CPU, but I have enough left to run a cabinet IR and some light effects.

@rotus8064 Год назад

hi. Is it like loading NAM from jack audio instead of DAW? If wio terminal is used only for display, can you tell me the jack audio setting?

@NostaticSoftware Год назад

There is nothing particularly special about my Jack settings, but I am running a custom Jack client with NAM integrated.

@dickjonesify Год назад

Nice! Now get building these for the rest of us NAM users 🤘🏼😆

@forrymight274 Год назад

Please tell me how you did it?

@outlawphil 6 месяцев назад

This seems more polished than the NeuralPi, how is the latency?

@archangelsean Год назад

How many ms of delay as that would concern me as not seen anything on linux that compares to macs coreaudio or windows asio etc

@gerrykavanagh Год назад

Using low-latency kernel, JackAudio and an AudioInjector card, I can get about 7ms round-trip using LV2 plugins on a Pi 3b. That's 7ms from hitting the string to hearing the sound.

@ZombieLincoln666 3 месяца назад

I’d be awesome if they could add this to Zynthian

@peterasberg9761 Год назад

How is the delay?

@NostaticSoftware Год назад

If you mean latency, it is very little. I'm able to run a 64-sample buffer.

@TheJabcob Год назад

How much ram needed on the Pi?

@NostaticSoftware Год назад

Depends on what else you've got going on with the Pi. Running NAM takes very little memory. I've got a 4gb Pi 4, but I'm guessing I could run on 1gb without issues.