Please ensure Javascript is enabled for purposes of website accessibility Jump to content

AI Technology


purplebeefunk
 Share

Recommended Posts

Hi Helix Team, 

 

As Spark 2 is coming to market with AI technology, it would be awesome if the NEXT GEN of Helix have the same feature as well.
I'm sure with all the goodies that Helix currently have, the AI feature will take it to another level 
Hope we can have the Helix 2.0 soon!!

Cheers

 

Link to comment
Share on other sites

I’m impressed with what AI can do in my very limited experience with it, so I’m curious. What specific features would you like AI to do/enable in the domain of guitar amp/FX processing?

Link to comment
Share on other sites

The best thing about AI at this point is by simply using those two letters in your marketing campaign you can potentially boost sales regardless of any difference it will make in the product's actual capabilities.

 

There is clearly vast potential in AI all of which has to be run on massive, very expensive computers that won't fit in your house much less in a modeling box.  It's also highly suspect when any company touts AI without qualifying what type of AI because they're all different.  The only type that really would make sense in the modeling world would be Generative AI which relies on the computer being taught how to use different components correctly within a signal chain to resolve to a specified sample sound.  Of course the accuracy of the sample sound would be suspect without knowing what kind of guitar setup was used to produce it.  And the sample itself could vary wildly depending on the settings of the guitar.  All of which can only happen by submitting each sample to a remote AI computer to be able to run it's program and determine the outcome.

 

My best guess what Spark may be referring to would be to use AI to design the individual modeled components from the real world you can use in your modeler.

 

But just imagine the level of bragging rights you'd have if you can tell people your modeler created the sound through AI even if it wasn't done that way.  Isn't marketing a great invention???!!!

  • Like 2
Link to comment
Share on other sites

The more guitarist can sound homogeneous the better.

 

As ai is only good at replicating existing forms and not capable of extending them the sooner it takes over tone the better and the quicker we will regress to the norm.

 

I welcome my tone overlords with welcome arms.

  • Haha 2
  • Upvote 1
Link to comment
Share on other sites

On 8/2/2024 at 5:01 AM, DunedinDragon said:

The best thing about AI at this point is by simply using those two letters in your marketing campaign you can potentially boost sales regardless of any difference it will make in the product's actual capabilities.

 

There is clearly vast potential in AI all of which has to be run on massive, very expensive computers that won't fit in your house much less in a modeling box.  It's also highly suspect when any company touts AI without qualifying what type of AI because they're all different.  The only type that really would make sense in the modeling world would be Generative AI which relies on the computer being taught how to use different components correctly within a signal chain to resolve to a specified sample sound.  Of course the accuracy of the sample sound would be suspect without knowing what kind of guitar setup was used to produce it.  And the sample itself could vary wildly depending on the settings of the guitar.  All of which can only happen by submitting each sample to a remote AI computer to be able to run it's program and determine the outcome.

 

My best guess what Spark may be referring to would be to use AI to design the individual modeled components from the real world you can use in your modeler.

 

But just imagine the level of bragging rights you'd have if you can tell people your modeler created the sound through AI even if it wasn't done that way.  Isn't marketing a great invention???!!!

 

Lol, indeed. AI, the new HD. And so much of it BS... but who wanted to risk being the last guy walking around without hi-def boxer shorts? ;). Madison Avenue has always loved a good fairy tale. Hang the right badge on your product, and you can sell snow to an Eskimo...but I digress. That being said, if AI shoelaces can tie themselves, sign me up. :)

  • Haha 1
Link to comment
Share on other sites

Few marketing blurbs differentiate between Machine Learning and Artificial Intelligence, so it's hard to know exactly what they're doing. For example:

 

  • Machine Learning could go through all the presets you use in live performance and balance their levels. It would "learn" what level each preset uses, compare them, and edit them for the same perceived value.
  • Artificial Intelligence could go through all the presets you use in live performance, and generate new presets which meet the same general criteria as what you seem to like. 

 

Companies are finding out that AI isn't generating much $$ yet, and there are also legal issues involved. AI has much potential -  I used it to generate two recent book covers. It could also be tremendously helpful in reducing repetitive work in the studio. But, so far, practical applications for music that truly use artificial intelligence don't seem to be very common. 

 

Presonally, I'm a fan of artificial stupidity. Some of my best sounds come from making mistakes and thinking "y'know, that actually sounds kind of cool."

 

 

  • Like 6
Link to comment
Share on other sites

Not for me.

I like having to figure things out for myself. It's how I know I'm alive.

I'm glad I won't be around in 50 years. I'd hate to have to deal with people who have grown up with AI - ie. 'the hard of thinking'.

Link to comment
Share on other sites

On 8/3/2024 at 9:00 AM, mlwinnig said:

AI EVERYTHING / EVERYWHERE!  Sounds like the next "Dot Bomb Bubble" to me.

 

So how are those self-driving semis, ride-shares and drone deliveries working out so far?

 

I supplement my income driving for Ubereats.

All day long I take driving directions from a robot.

If I paid attention I'd be dead.

  • Like 1
  • Haha 2
Link to comment
Share on other sites

On 8/3/2024 at 5:32 AM, somebodyelse said:

I like having to figure things out for myself.

 

Bingo. I can't think of anything I'd want less than an AI thingie that could generate guitar parts for me. Creating guitar parts makes me happy. So does creating Helix presets. 

  • Upvote 1
Link to comment
Share on other sites

On 8/3/2024 at 2:03 AM, craiganderton said:

Presonally, I'm a fan of artificial stupidity. Some of my best sounds come from making mistakes and thinking "y'know, that actually sounds kind of cool."

 

 

 

Nothing "artificial" about that. That's old school stupid... as real as it gets. We've all been there, lol. Bob Ross used to call them "happy accidents"...;)

  • Haha 1
Link to comment
Share on other sites

Anyone who has used Tonex, Sonible plug-ins, Chat GPT, or AI art-generation such as Dali surely must realize there is an enormous amount of power and potential in AI (machine learning, neural networks, etc). Unfortunately, a lot of companies are claiming AI features as a buzzword without having any real substance.

  • Like 1
Link to comment
Share on other sites

On 8/2/2024 at 11:03 PM, craiganderton said:

Few marketing blurbs differentiate between Machine Learning and Artificial Intelligence, so it's hard to know exactly what they're doing. For example:

 

  • Machine Learning could go through all the presets you use in live performance and balance their levels. It would "learn" what level each preset uses, compare them, and edit them for the same perceived value.
  • Artificial Intelligence could go through all the presets you use in live performance, and generate new presets which meet the same general criteria as what you seem to like. 

 

Companies are finding out that AI isn't generating much $$ yet, and there are also legal issues involved. AI has much potential -  I used it to generate two recent book covers. It could also be tremendously helpful in reducing repetitive work in the studio. But, so far, practical applications for music that truly use artificial intelligence don't seem to be very common. 

 

Presonally, I'm a fan of artificial stupidity. Some of my best sounds come from making mistakes and thinking "y'know, that actually sounds kind of cool."

 

 

 Man, I would LOVE AI auto-volume leveling between patches, or even within the same one with different guitars or even just coil splitting pickups.  Some sort of algorithm that can set a minimum volume for when you back off pick attack but still is full volume (up to your predetermined maximum) when you dig in a little more or strum hard but without the squash of a compressor.  Would make it much easer for multiple guitar gigs.  Sign me up!

 

Or even something as simple as how Kempers profile.. just put a bunch of noise through and set output volumes on all your patches accordingly.  That would be sweet!  Probably much less CPU intensive as well.

Link to comment
Share on other sites

  • 2 weeks later...
On 8/2/2024 at 4:01 AM, DunedinDragon said:

There is clearly vast potential in AI all of which has to be run on massive, very expensive computers that won't fit in your house much less in a modeling box. 

This really isn't a limitation.  All commercial generative AI services offer API's which could easily be integrated into a flagship level product.  This would require the pedal have a wi-fi connection or network connection of some sort.  Or if not the pedal itself, API connectivity through the editing software.  But you certainly wouldn't need to have the AI engine in the pedal itself.

Link to comment
Share on other sites

On 8/16/2024 at 11:24 AM, cglaser68 said:

This really isn't a limitation.  All commercial generative AI services offer API's which could easily be integrated into a flagship level product.  This would require the pedal have a wi-fi connection or network connection of some sort.  Or if not the pedal itself, API connectivity through the editing software.  But you certainly wouldn't need to have the AI engine in the pedal itself.

That's probably reasonably close to what the first generation of something like this might look like.  But the only way for it to have any real value to the user would require some personalization in the training of the AI as to how his system is organized (type of guitar, signal output setup such as amps, direct to PA, etc) as well as personal preferences for the type of sound desired for each individual user as those are so unique in each situation because it's typically based on an emotional response to the result which AI is struggling to figure out how to do.  Who knows what the pricing model for such a thing might be.

  • Upvote 1
Link to comment
Share on other sites

  • 1 month later...
On 8/2/2024 at 5:45 AM, purplebeefunk said:

Hi Helix Team, 

 

As Spark 2 is coming to market with AI technology, it would be awesome if the NEXT GEN of Helix have the same feature as well.
I'm sure with all the goodies that Helix currently have, the AI feature will take it to another level 

 


Hope we can have the Helix 2.0 soon!! Thanks to the technologies offered by AI today, it is possible to achieve success in many areas. In particular, fintech development, which you can read about here https://dashdevs.com/fintech-software-development/ and learn more about this topic. The potential of AI is hard to overestimate, as well as to predict how it will develop in the near future.

Cheers

 

It would definitely be exciting to see AI features integrated into the next generation of Helix! With AI becoming a huge part of tech advancements, having it in Helix 2.0 could really enhance its capabilities and take it to the next level. Spark 2's move toward AI shows how important it is to stay ahead in this space. Combining Helix's current features with AI could mean more intuitive, personalized experiences, making it a real game changer.

Link to comment
Share on other sites

On 10/14/2024 at 11:18 AM, silverhead said:

I’m curious. What specific features or capabilities do you imagine AI could provide?

 

We could get it to admonish you whenever you hit a wrong note...bonus points if they include a retractable wooden ruler that smacks you on the knuckles for that vintage Catholic school nun vibe...;)

  • Haha 1
Link to comment
Share on other sites

I remember the Boss GT series had an EZ Tone Creator.   You would move a point on an X/Y axis chart and as you moved around it would, in the background, adjust parameters in amp selection/settings, as well as some potential effects like Delay.  X axis was rhythm (far left) and lead (far right), the Y axis was Heavy Rock at the top and Clean at the bottom.  So you could float around that chart and then it would adjust your amp/cab selections.   The next chart you could choose would be an FX based one sorta thing.  


It was pretty cool to choose your genre, work through the 2 charts, and then save the preset and look at the nuts and bolts of what their engine drew up.  You'd end up on some amps that normally you'd just not try out for whatever reason.    Was a great way to get some really quick base tones....would love to see Helix do this as well.  

Link to comment
Share on other sites

Today's Generative AI and Machine Learning is mostly recreating things that already existed as captured in their training sets. That's great for making this content more readily available and reusable, pending IP rights issues. The AI models can possibly produce some interesting new results based on validation errors and hallucinations. But AI can't currently create Line 6 Litigator, one of the best amp models in Helix. 

 

So an interesting question to explore might be: How can AI help someone like Ben Adrian be more creative, innovative and productive to produce the next new products that could be fed into future training sets to progress the industry?

 

Or will AI just reduce the need for creative engineering while the product owners maximize profits from products of the past, saturating the market with the same things over and over?

 

 

  • Thanks 2
Link to comment
Share on other sites

Personally I think most people overestimate the value of what generative AI can do when compared to actual human intellectual capabilities.  Ultimately it depends on how well it's been trained in order to consider "out of the box" solutions to problems.  Someone like Ben Adrian is unique in that regard so a system trained by Ben would likely exhibit more of those traits than one trained by a very strict "rules based" individual.  I think we're quite a few years away from understanding how creativity works in humans before we'll be able to even approach considering how to design it into AI computers.

Link to comment
Share on other sites

Join the conversation

You can post now and register later. If you have an account, sign in now to post with your account.
Note: Your post will require moderator approval before it will be visible.

Guest
Reply to this topic...

×   Pasted as rich text.   Paste as plain text instead

  Only 75 emoji are allowed.

×   Your link has been automatically embedded.   Display as a link instead

×   Your previous content has been restored.   Clear editor

×   You cannot paste images directly. Upload or insert images from URL.

 Share

×
×
  • Create New...