<It’s funny, but I seem to detect just a tiny bit of vocal cracking at the beginning of the cha-a-a-a-nce bit, as if Roy was overstretching his natural range>
It's more of a crude vibrato, but that works in Roy's favor, especially when the payoff is the hillbilly twang that saturates *YOOOOOOU'VE GOT TO TAKE* adding a savory soulfulness to the melodrama.
<Prepare yourself for some tight bonding with the word ‘blue’, which, judging by the frequency of its appearance in Roy’s lyrics, must have been his favorite color>
Much like his spiritual successor/fanboy Jeff Lynne, ie Bluebird is Dead, Boy Blue, Mr Blue Sky, Midnight Blue, Birmingham Blues, etc....
<Roy’s lead and corny-and-even-cornier backing vocals (instead of dum-dum-dum-dumdy-doo-wah, we now have sha-la-la-dooby-wah, dum-dum-dum-yeh-yeh-um, which is definitely more sophisticated but not necessarily «transcendentally progressive»>
Yep, yep, um. Is there such a thing as Dumb Complexity? Simple Profundity? Fluent Baby Talk? The dum-dum-dumness of those bgvs are supposed to represent a kind of moon-eyed besotted jibberish of young lovers, I think. But then he hits those hillbilly crescendos and we get the get the heavenly ecstasy of the tongues of angels.
Wow. Interesting take, George. Thought it possible that some of us would take a more equivocal point of view on AI in music than my instinctive condemnation. As I've replied to umpteen such videos posted, "Why? What's the point???!!!". I still don't get it, except as an ephemeral fad quickly put behing us. But if some of you think it might be what we deserve, and that it's just a bit of good fun with improbable positive benefits, I'm happy to play along -- for another week or two (sigh).
So funny how AI is already becoming the central talking point of any discussion these days. I went to a presentation of Mahler's 4th Sympnoy last night and the opening screen image was one of Mahler's head partly interwoven with various associative elements of his life. I thought to myself, yes, that's very well done, i'll have to ask the presenter who was responsible for the striking image. But before I could do so, he revealed he had asked ChatGPT to come up with something. No real sense of irony in his admission, as if it was already an accepted way of doing things. As is, of course, the assinine current wave of AI cover songs (e.g., Lennon & McCartney trading verses on Radiohead's Karma Police), which looks well set to become more than just a passing fad. i'm guessing Roy Orbison's unique vocal timbre and singing style will allow him to escape unscathed, but I wouldn't bet on it!
In one of Rick Beato's last videos, he made a good point that the grounds for the "AI takeover" had actually been laid out quite nicely in the previous decades - with human-made pop music starting to sound more and more like it was made by computers in the first place (everything from Autotune to dynamic compression, etc.), people are now more than ready to accept actual computers moving in. In fact, these Kurt Cobain and "young McCartney" vocals do sound more human-like than quite a bit of the modern pop productions, so I can understand the excitement and enthusiasm for these "Brides of Frankenstein". It's kind of creepy to try and imagine the artistic landscape ten years from now, but hey, maybe if the reanimated ghost of John Lennon is in more demand these days than actual young musicians, we deserve it.
Totally agree. With how overprocessed and inorganic a lot of music sounds these days, I feel that an AI could write a lot of music that is just as good and nobody would tell the difference. An AI would have to be rather complex to recreate the unconventional grit of many bluesmen or rock musicians, but with a lot of neo-soul and pop with very dense electronic production and autotune that relies on one or two musical phrases easily could be recreated. Though, the frustrating part is that everybody is so brainwashed by the "glory of AI" that nobody is even properly thinking about whether such a musical landscape is one they will feel comfortable with. Though, even "idol" culture seems to have a loosening grip as well, so maybe people will go "oh who cares? A catchy tune is a catchy tune." Time will tell I guess.
The philosophical conundrum here is that the so-called AI, creating "new" music, does not really "know" or "understand" (let alone "feel") whatever it is creating; it uses its huge banks of data to emulate the form without having either an innate or an educated comprehension of the substance. The machine can grasp the concept of syntax, but not that of semantics.
Theoretically, it is possible for it to accidentally fall upon something "substantial" the same way a toddler could accidentally compose a meaningful sentence from a bunch of alphabet blocks. But the sad thing is that (judging primarily by comments left on these videos by YouTubers) for many people, "form" alone would be fully sufficient to satisfy their needs, just the same way they are happy to consume modern pop garbage as the background soundtrack for their everyday life.
The thing is that I think a lot of people don't even realize that it is just the "form" that they enjoy: rather, that the mere recreation of form and seeming serious might actually seem emotional and moving. At least from what I've noticed, some people switching to music of actual substance (which I guess is a somewhat subjective term, but I think we can draw a line somewhere on that) either provokes a strong negative or positive reaction, almost as if they were a prisoner who just got released who either embraces freedom or doesn't even know how to process and manage it. Which is why I think that the problem is even deeper than that because people will honestly be able to reason through it being just as "emotional" and "serious" (especially since I've met some people who really believe AI is like the savior of mankind). Anyways I'll stop sending so many long winded replies on this thread so this doesn't become a massive clusterfuck of conversation.
It feels like people’s brains have been rewired such that music by itself is not a good enough stimulus anymore to keep them occupied. The information age basically spells the death of any music which is not background noise on some level. A skilled information consumer will adjust the level of complexity in music to accompany his other cognitive activities, but music by itself can never capture all his attention. This has degraded people’s ability to enjoy music on its own terms and inevitably changes the nature of music being produced and listened to.
" In other words, you could probably build an AI that would, more or less correctly, predict post-1960 Roy Orbison if you fed Sings Lonely And Blue into it, but you certainly couldn’t do the same for the Beatles if you only fed it Please Please Me and its surrounding singles."
I agree with your general point, but this is selling Roy way too short. There's no way an AI would somehow spit out things like Running Scared or Oh Pretty Woman. Roy is not *that* predictable that an AI could literally build his whole output in his prime years.
Well, to be fair, what is currently being peddled as "AI" couldn't truly spit out a genuine AC/DC song, let alone a Roy Orbison. So it's a hyperbole all the way anyway, but I'm fine with just the general point.
<It’s funny, but I seem to detect just a tiny bit of vocal cracking at the beginning of the cha-a-a-a-nce bit, as if Roy was overstretching his natural range>
It's more of a crude vibrato, but that works in Roy's favor, especially when the payoff is the hillbilly twang that saturates *YOOOOOOU'VE GOT TO TAKE* adding a savory soulfulness to the melodrama.
<Prepare yourself for some tight bonding with the word ‘blue’, which, judging by the frequency of its appearance in Roy’s lyrics, must have been his favorite color>
Much like his spiritual successor/fanboy Jeff Lynne, ie Bluebird is Dead, Boy Blue, Mr Blue Sky, Midnight Blue, Birmingham Blues, etc....
<Roy’s lead and corny-and-even-cornier backing vocals (instead of dum-dum-dum-dumdy-doo-wah, we now have sha-la-la-dooby-wah, dum-dum-dum-yeh-yeh-um, which is definitely more sophisticated but not necessarily «transcendentally progressive»>
Yep, yep, um. Is there such a thing as Dumb Complexity? Simple Profundity? Fluent Baby Talk? The dum-dum-dumness of those bgvs are supposed to represent a kind of moon-eyed besotted jibberish of young lovers, I think. But then he hits those hillbilly crescendos and we get the get the heavenly ecstasy of the tongues of angels.
Wow. Interesting take, George. Thought it possible that some of us would take a more equivocal point of view on AI in music than my instinctive condemnation. As I've replied to umpteen such videos posted, "Why? What's the point???!!!". I still don't get it, except as an ephemeral fad quickly put behing us. But if some of you think it might be what we deserve, and that it's just a bit of good fun with improbable positive benefits, I'm happy to play along -- for another week or two (sigh).
So funny how AI is already becoming the central talking point of any discussion these days. I went to a presentation of Mahler's 4th Sympnoy last night and the opening screen image was one of Mahler's head partly interwoven with various associative elements of his life. I thought to myself, yes, that's very well done, i'll have to ask the presenter who was responsible for the striking image. But before I could do so, he revealed he had asked ChatGPT to come up with something. No real sense of irony in his admission, as if it was already an accepted way of doing things. As is, of course, the assinine current wave of AI cover songs (e.g., Lennon & McCartney trading verses on Radiohead's Karma Police), which looks well set to become more than just a passing fad. i'm guessing Roy Orbison's unique vocal timbre and singing style will allow him to escape unscathed, but I wouldn't bet on it!
In one of Rick Beato's last videos, he made a good point that the grounds for the "AI takeover" had actually been laid out quite nicely in the previous decades - with human-made pop music starting to sound more and more like it was made by computers in the first place (everything from Autotune to dynamic compression, etc.), people are now more than ready to accept actual computers moving in. In fact, these Kurt Cobain and "young McCartney" vocals do sound more human-like than quite a bit of the modern pop productions, so I can understand the excitement and enthusiasm for these "Brides of Frankenstein". It's kind of creepy to try and imagine the artistic landscape ten years from now, but hey, maybe if the reanimated ghost of John Lennon is in more demand these days than actual young musicians, we deserve it.
Totally agree. With how overprocessed and inorganic a lot of music sounds these days, I feel that an AI could write a lot of music that is just as good and nobody would tell the difference. An AI would have to be rather complex to recreate the unconventional grit of many bluesmen or rock musicians, but with a lot of neo-soul and pop with very dense electronic production and autotune that relies on one or two musical phrases easily could be recreated. Though, the frustrating part is that everybody is so brainwashed by the "glory of AI" that nobody is even properly thinking about whether such a musical landscape is one they will feel comfortable with. Though, even "idol" culture seems to have a loosening grip as well, so maybe people will go "oh who cares? A catchy tune is a catchy tune." Time will tell I guess.
The philosophical conundrum here is that the so-called AI, creating "new" music, does not really "know" or "understand" (let alone "feel") whatever it is creating; it uses its huge banks of data to emulate the form without having either an innate or an educated comprehension of the substance. The machine can grasp the concept of syntax, but not that of semantics.
Theoretically, it is possible for it to accidentally fall upon something "substantial" the same way a toddler could accidentally compose a meaningful sentence from a bunch of alphabet blocks. But the sad thing is that (judging primarily by comments left on these videos by YouTubers) for many people, "form" alone would be fully sufficient to satisfy their needs, just the same way they are happy to consume modern pop garbage as the background soundtrack for their everyday life.
The thing is that I think a lot of people don't even realize that it is just the "form" that they enjoy: rather, that the mere recreation of form and seeming serious might actually seem emotional and moving. At least from what I've noticed, some people switching to music of actual substance (which I guess is a somewhat subjective term, but I think we can draw a line somewhere on that) either provokes a strong negative or positive reaction, almost as if they were a prisoner who just got released who either embraces freedom or doesn't even know how to process and manage it. Which is why I think that the problem is even deeper than that because people will honestly be able to reason through it being just as "emotional" and "serious" (especially since I've met some people who really believe AI is like the savior of mankind). Anyways I'll stop sending so many long winded replies on this thread so this doesn't become a massive clusterfuck of conversation.
It feels like people’s brains have been rewired such that music by itself is not a good enough stimulus anymore to keep them occupied. The information age basically spells the death of any music which is not background noise on some level. A skilled information consumer will adjust the level of complexity in music to accompany his other cognitive activities, but music by itself can never capture all his attention. This has degraded people’s ability to enjoy music on its own terms and inevitably changes the nature of music being produced and listened to.
" In other words, you could probably build an AI that would, more or less correctly, predict post-1960 Roy Orbison if you fed Sings Lonely And Blue into it, but you certainly couldn’t do the same for the Beatles if you only fed it Please Please Me and its surrounding singles."
I agree with your general point, but this is selling Roy way too short. There's no way an AI would somehow spit out things like Running Scared or Oh Pretty Woman. Roy is not *that* predictable that an AI could literally build his whole output in his prime years.
Well, to be fair, what is currently being peddled as "AI" couldn't truly spit out a genuine AC/DC song, let alone a Roy Orbison. So it's a hyperbole all the way anyway, but I'm fine with just the general point.