Hacker News Board Hacker News, browsed like a classic message board
Live data from Hacker News

Show HN: SmartGuitarAmp – Guitar plugin made with deep learning

github.com

Post reply on HN
Print view View original

71–80 of 163 posts

Re: Show HN: SmartGuitarAmp – Guitar plugin made with deep learning

#71

Post by TylerE »

bjelkeman-again wrote: post #68
TylerE wrote: post #58

Earlier quoted context omitted.

That's unlikely to end well. You really need custom audio hardware - or at least dedicated commercial-grade stuff, for this sort of application. Latency is HUGELY important. 2-3ms can be felt, and most consumer level stuff uses buffers WAY bigger than that.

As a reference the new kid on the block, NeuralDSP’s Quad Cortex uses 500 Mhz Quad-Core SHARC https://www.analog.com/en/products/processors-microcontrolle...

Yes, SHARC is pretty widespread.

I'm not talking about the DSP though.

You need an ADC on the frontend and a DAC on the backend.

Re: Show HN: SmartGuitarAmp – Guitar plugin made with deep learning

#72

Post by hluska »

dang wrote: post #41
keyth72 wrote: post #29

Earlier quoted context omitted.

Ah, thank you for pointing that out. So if I wanted to use the title I have, I should write an article with that title. But even then it seems I wasn't in line with the quality of posts expected for this site. Will definitely keep that in mind for future posts.

I'm a mod here and have changed the title to one that fits the guidelines better. Don't worry—as others have said, it's a great post! I hope you'll continue to share your work on HN and maybe participate in other discussions you find interesting. (For those who are curious, the submitted title was "Free guitar plugin on GitHub sounds like $600 tube amp by using deep learning".)

This is an unbelievably kind response. Nice job friend and thanks for working on a Saturday!!! :)

Re: Show HN: SmartGuitarAmp – Guitar plugin made with deep learning

#73

Post by hluska »

keyth72 wrote: post #48
whiddershins wrote: post #36

Earlier quoted context omitted.

Don’t let any of this discourage you tho. This is a great contribution to the community. This site just has a preference to more technically descriptive titles.

Well I'm an engineer so that shouldn't be too hard, usually people want me to dumb it down! Considering this project sat for a month with no stars and after this post (and one hour) it jumped to 134 stars, not discouraged at all!

You know friend, I haven’t tried your project (yet) but your documentation is excellent as are your responses to all the comments on this thread. I think you have a whole lot to be proud of and I haven’t even tried your plugin yet!!

Re: Show HN: SmartGuitarAmp – Guitar plugin made with deep learning

#76

Post by offtop5 »

TylerE wrote: post #71
bjelkeman-again wrote: post #68

Earlier quoted context omitted.

As a reference the new kid on the block, NeuralDSP’s Quad Cortex uses 500 Mhz Quad-Core SHARC https://www.analog.com/en/products/processors-microcontrolle...

Yes, SHARC is pretty widespread. I'm not talking about the DSP though. You need an ADC on the frontend and a DAC on the backend.

Thanks for the input. If I had 500$ or so could I make a digital bass pedal / tube amp ,?

Re: Show HN: SmartGuitarAmp – Guitar plugin made with deep learning

#77

Post by dmacvicar »

dyeje wrote: post #61
w0mbat wrote: post #44

For people who don’t play guitar, this is really commonplace technology now. Engineers worked out how to model the circuit behavior of any amp a while ago. Powerful valve guitar amps have really interesting effects on the waveform particularly when driven hard, but that valve technology is expensive, unreliable, bulky, and very loud when producing some of the most sought after tones. The most popular guitar amps now…

> The most popular guitar amps now are all solid state Maybe on the low end of the market, but tube amps still reign supreme for most serious players.

Many prominent artists are touring with modelling amps for years now. https://www.fractalaudio.com/artists/

There are many advantages. Less maintenance, lower stage volumes, better for in-ear-monitoring use, sound consistency, less weight to carry, and the ability to explore hundred of amps sounds with a single device eg. check https://wiki.fractalaudio.com/wiki/index.php?title=Amplifier...

Re: Show HN: SmartGuitarAmp – Guitar plugin made with deep learning

#78

Post by TylerE »

offtop5 wrote: post #76
TylerE wrote: post #71

Earlier quoted context omitted.

Yes, SHARC is pretty widespread. I'm not talking about the DSP though. You need an ADC on the frontend and a DAC on the backend.

Thanks for the input. If I had 500$ or so could I make a digital bass pedal / tube amp ,?

Make, probably not..

But at that price point just get a used Helix LT which will be light years ahead of anything you can bodge together.

This is one of those areas where both the hardware AND the software really matter.

You can't touch the available hardware because of economies of scale - they're buying DSP chips, enclosures, power supplies, by the thousand at least.

And they have multi-person full time teams working on the software.

Re: Show HN: SmartGuitarAmp – Guitar plugin made with deep learning

#79

Post by dmacvicar »

nom wrote: post #46

Awesome project and pretty good results from a mathematical standpoint. The biggest challenge is to try and convince the tube amp enthusiasts that it sounds indistinguishable to the real hardware ;)

Most people, including guitar players, can't tell the difference between popular modelling devices and real tube amps when subject to a real blind-test.

The of neural networks in this field should be measured against current modelling technology, not real hardware.

Re: Show HN: SmartGuitarAmp – Guitar plugin made with deep learning

#80

Post by sickcodebruh »

bjelkeman-again wrote: post #64
keyth72 wrote: post #12

Earlier quoted context omitted.

Thank you! Any feedback is appreciated if you get around to trying it out. If you look at the WaveNetVA repository from Aalto University that my plugin is based on, at least one person affiliated with that project works at Neural DSP. I suspect that the QuadCortex uses something like this (maybe a more advanced neural net) combined with other proprietary techniques and application specific hardware. They claim they c…

It is really interesting to see this becoming more widespread. I am waiting for a delivery in the first batch of the Quad Cortex.

If real world reviews confirm the previews I’ve encountered, that’ll be the one for me. It really seems phenomenal.
Post reply on HN