AI assisted blink remixes (Mark singing Tom parts, Tom singing Mark's, BCR songs in the style of blink, modern songs in the old style, etc. etc.)

[M!ke]

I find this entire subject very interesting, and I'm sure that in time we're only going to see more and more of these with more and more convincing results (some of these are already pretty convincing).  This probably has been discussed in recent months in the general any topic, but again, given that I think that we'll see only more and more of this in time, it might be nice to have a thread dedicated to it and we can all just post whatever of these AI mixes that we enjoy from the likes of youtube.

Really really enjoy this version of There Is if blink had performed it, pretty cool to hear (seemingly) Mark and Tom trade off vocals in such a song:

To compliment a Tom project featuring AI Mark, here is an alright one for a Mark project featuring AI Tom, No It Isn't:

Tom sings Adam's Song (this one is really solid):

Mark sings First Date

Tom sings The Rock Show

Mark and Tom swap vocals in Elevator (this one is very strong):

Baby Come On featuring Tom (kinda falls apart toward the end, but otherwise I really like this one too):

Edging in the oldschool Tom style (clip, not full song) (and actually after looking at this one again, I'm not so sure it was AI (though it did come up when I searched for blink 182 AI), still it's interesting all the same, so I'll keep it here):

There are plenty of others on Youtube, but be warned, a lot of them are way too obviously AI to be very enjoyable (much more so than the ones I've presented here).  But yeah, if you find any that you enjoy, feel free to post them, you'll note that most of these are from very recently, like within the last month or so, if that.  I think it's fair to say that we're in store for plenty more of this kind of stuff.

[just_bringing_good_chats]

thx for this post, some of those re-mixes are very good and or very interesting

 

after all mark sang on a boxcar raccer song, even in the midst of that pseudo feud in between him and tom,

so not far fetched that he could of been on other songs, and even just part of that band(not saying he should of, it was nice for tom to go do his own thing seperatly, that was something very diff than blink)

[The Watcher on the Wall]

The There Is one is pretty cool musically, but it’s hard to take it seriously with “Mark” pronouncing words like Tom 🤣

[M!ke]

4 minutes ago, The Watcher on the Wall said:

The There Is one is pretty cool musically, but it’s hard to take it seriously with “Mark” pronouncing words like Tom 🤣

Yeah, that's just where the AI is currently at for music really.  I imagine that within only a year or so that this stuff will only be all the more sophisticated and convincing.

[The Watcher on the Wall]

43 minutes ago, M!ke said:

Yeah, that's just where the AI is currently at for music really.  I imagine that within only a year or so that this stuff will only be all the more sophisticated and convincing.

This might be a really dumb question, but is the AI doing the instruments on that or did someone make a cover and add the AI vocals?

[JarJarBlinks]

I hate this shit

[Worst poster here]

I watched a few and AI can replicate Mark a lot better than Tom. I think I’d actually be able to know the Tom parts were AI even if I went in blind. 

[Smooth My Cum]

This is pretty rad @M!ke. Thanks for posting.

 

Just watched There Is and wasn’t disappointed. Moving onto the rest 

[Smooth My Cum]

I wonder how cool it would be to convert your own voice to Tom or Mark’s. Could probably make a blink song of your own. I can see somebody posting AvA songs as well in the style of 1999-2004 Tom also

[Matt.]

At least all these shitty videos are contained in one place I can easily avoid!

[M!ke]

1 hour ago, The Watcher on the Wall said:

This might be a really dumb question, but is the AI doing the instruments on that or did someone make a cover and add the AI vocals?

Not entirely sure.  Pretty sure AI can do instrumentation these days, but I dunno if that's what's happening or not in any of these.

[Smooth My Cum]

The instrumentals on “There Is” are really close to blink’s regarding the tone. Usually when you see somebody on YouTube make a blink sounding song of their own, the tone is off and it’s noticeable. I have a feeling they remixed existing tracks or they used AI.  
 

Would be sick actually if an AI could create a virtual guitar rig that mimics the tone of somebody else.  Basically, record the direct input of your guitar, throw in the target sound you want it to be, and then have it create a virtual guitar effects chain.

 

The What’s My Age Again tone would be my first go-to for that. That tone is really iconic and they never really reused it. And the drone guitar from Everytime I Look for You would make for some sick octave riffs.

[Cum Back Tom]

This feels dirty and like I can't look away.  It's like the computers are taking me over and this is their last stop...stop, help me get GET AWAY FROM ME GET THAT ROBOT ARM AWAY FROM MY ASSHOLE, ASSHOLE

 

There Is is pretty good.

[Champ182]

12 hours ago, Smooth My Balls said:

Would be sick actually if an AI could create a virtual guitar rig that mimics the tone of somebody else.  Basically, record the direct input of your guitar, throw in the target sound you want it to be, and then have it create a virtual guitar effects chain.

The What’s My Age Again tone would be my first go-to for that. That tone is really iconic and they never really reused it. And the drone guitar from Everytime I Look for You would make for some sick octave riffs.

I'm not well-read in this world at all, but I THINK that's sort of how the new digital rigs that Tom and Mark use work... They're able to "read" specific real amp tones that Jerry Finn used on the old albums and an algorithm tests and retests and retests 1000x until it puts out a very convincing recreation of that specific tone. Mark's bass tech explained it on reddit recently, I'll try to find it.

EDIT: Here is his post:

"Hi everyone. I am Mark's bass tech, Brian Diaz. Just to clarify what this is...

I was given some profile files of a bass amp (Ampeg SVT) that was once owned by Jerry Finn made by a producer in the Bay Area who now owns it. What this means is that in a studio setting the amp and speakers were mic'd up and "captured" for a digital modeling amp - in this case a Kemper and Neural Quad Cortex. Live Mark is going to be using a Neural QC for his bass sounds (formerly he was using a Kemper for many years. In my personal opinion I think the QC completely obliterates Kemper in terms of sound and in ease of use) as he's discussed in his Discord and Twitch. It is a very realistic sounding replication of a tube amplifier using the capture made of the original amp. It involves some wildly complex algorithms, but ultimately it's meant to be a very convincing duplication of the characteristics of the original but in a very portable and stable digital format. We've spent time in studio and rehearsal tweaking these sounds to emulate classic album tones and current sounds used on the newest albums.

TL;DR explanation: it sounds fucking sick!

Hope that clears up any confusion as to what it means when the "Jerry Finn SVT" is mentioned! The technology isn't necessarily new but it's gotten way better over the last decade."

Also, if you search his posts he gives out a bunch of really cool info about this kind of stuff: https://www.reddit.com/user/bkd1976

[Smooth My Cum]

6 hours ago, Champ182 said:

I'm not well-read in this world at all, but I THINK that's sort of how the new digital rigs that Tom and Mark use work... They're able to "read" specific real amp tones that Jerry Finn used on the old albums and an algorithm tests and retests and retests 1000x until it puts out a very convincing recreation of that specific tone. Mark's bass tech explained it on reddit recently, I'll try to find it.

EDIT: Here is his post:

"Hi everyone. I am Mark's bass tech, Brian Diaz. Just to clarify what this is...

I was given some profile files of a bass amp (Ampeg SVT) that was once owned by Jerry Finn made by a producer in the Bay Area who now owns it. What this means is that in a studio setting the amp and speakers were mic'd up and "captured" for a digital modeling amp - in this case a Kemper and Neural Quad Cortex. Live Mark is going to be using a Neural QC for his bass sounds (formerly he was using a Kemper for many years. In my personal opinion I think the QC completely obliterates Kemper in terms of sound and in ease of use) as he's discussed in his Discord and Twitch. It is a very realistic sounding replication of a tube amplifier using the capture made of the original amp. It involves some wildly complex algorithms, but ultimately it's meant to be a very convincing duplication of the characteristics of the original but in a very portable and stable digital format. We've spent time in studio and rehearsal tweaking these sounds to emulate classic album tones and current sounds used on the newest albums.

TL;DR explanation: it sounds fucking sick!

Hope that clears up any confusion as to what it means when the "Jerry Finn SVT" is mentioned! The technology isn't necessarily new but it's gotten way better over the last decade."

Also, if you search his posts he gives out a bunch of really cool info about this kind of stuff: https://www.reddit.com/user/bkd1976

That’s fucking sick.  I’ve been considering investing $1000 or so into some new guitar gear… but with AI and everything ramping up, I wonder what’s going to be hitting the market in the next couple years

[Jon Benjamin]

First date and the rock show were the best executions 

[Rickay]

To me they all sound like someone doing impressions of blink

[Fap God]

Yea, these AIs are stupid. 

[Smooth My Cum]

@Champ182 so I did some research into this kind of stuff today.  I landed on Tonex and AI Machine Modeling.

 

Looks like AI has been in the mix over the years, but this one actually has some really good tones.

 

I’m trying to find something where I can record the direct input of my guitar and have it run an algorithm to match the tone of a blink guitar sample. (Tonex may let you do something like that - some of the videos made it seem like you could train a model on your gear so that you can virtualize it.  Trying to find something that can train off of something like an mp3 of “Anthem” or “What’s My Age Again?”.  I may have to ultimately trick something into believing the sound from my rig is producing those samples.)