Life's inevitable afflictions: death, taxes, and public taste.
You thought Woodduck was critical LOL. This is priceless
1 - 20 of 25 Posts
I worry about AI. We are already seeing AI produced music by the likes of Oasis and the Beatles. How long before they turn to the classics and we start to see new symphonies by Beethoven? And will anyone actually know the difference?Life's inevitable afflictions: death, taxes, and public taste.
Just the other day I was amazed to listen to an AI-produced rendition of Yesterday by Freddie Mercury.I worry about AI. We are already seeing AI produced music by the likes of Oasis and the Beatles. How long before they turn to the classics and we start to see new symphonies by Beethoven? And will anyone actually know the difference?
It does sound like his voice, but it doesn't do any of the things that no doubt Mercury would have done, none of his vocal show-off.
And this makes me think that classical music, where variations are more restricted and predictable in the case they occur, is a more fertile ground for this kind of experiment, which, with technology's upgrading, may result in difficult to spot forgeries.
Not to worry. None of us will be around to witness the frightening state of a new existence. I sure would not want to raise my children in an atmosphere of an artificial world where opera will someday be looked upon as a charming relic of a gone era.I worry about AI. We are already seeing AI produced music by the likes of Oasis and the Beatles. How long before they turn to the classics and we start to see new symphonies by Beethoven? And will anyone actually know the difference?
Has anyone watched Futurama? Calculon, the robot, is a leading movie star there.
I believe we'll know the difference. Or the experts will. And by experts I mean many of us, without academic qualifications but a lifetime of listening to the real thing. AI can't provide soul or suffering, however technically astute.I worry about AI. We are already seeing AI produced music by the likes of Oasis and the Beatles. How long before they turn to the classics and we start to see new symphonies by Beethoven? And will anyone actually know the difference?
I wish I shared your optimism. "We" might, but "we are probably in a minority. The majority can't hear the difference between Russell Watson and Pavarotti, or Katherine Jenkins and Renée Fleming.I believe we'll know the difference. Or the experts will. And by experts I mean many of us, without academic qualifications but a lifetime of listening to the real thing. AI can't provide soul or suffering, however technically astute.
What will be nice is if they can use AI to improve the sound of early opera recordings. Now that is where the research should be focused. It is not perfect. You would not believe what AI thinks I say. I was writing to a friend to tell her I looked forward to seeing what her pastries were like and it wrote out " I look forward to seeing what her panties were like".
Beethoven who?I worry about AI. We are already seeing AI produced music by the likes of Oasis and the Beatles. How long before they turn to the classics and we start to see new symphonies by Beethoven? And will anyone actually know the difference?
A common experience, I gather, in texting. My peeve is having my computer trying to complete words and sentences for me. It happens here constantly, and I'm constantly impressed by the AS (Artificial Stupidity) of it. I haven't figured out how to make it stop; when I ask Google, AI does not step in and answer the question.What will be nice is if they can use AI to improve the sound of early opera recordings. Now that is where the research should be focused. It is not perfect. You would not believe what AI thinks I say. I was writing to a friend to tell her I looked forward to seeing what her pastries were like and it wrote out " I look forward to seeing what her panties were like".
You’ve heard of edible panties, right?What will be nice is if they can use AI to improve the sound of early opera recordings. Now that is where the research should be focused. It is not perfect. You would not believe what AI thinks I say. I was writing to a friend to tell her I looked forward to seeing what her pastries were like and it wrote out " I look forward to seeing what her panties were like".
Mas, I try not to think of such things!!!!!You’ve heard of edible panties, right?
You can't do much to improve the sound when the problem is what was not captured in the recording, i.e. the fundamental problem being the very restricted frequency and dynamic ranges of early recordings. Anything you do requires making assumptions about what might have been lost which, of course, you can never really know. You can try reducing noise but that isn't possible without the high likelihood of also removing some of what you actually want to keep.What will be nice is if they can use AI to improve the sound of early opera recordings. Now that is where the research should be focused. It is not perfect. You would not believe what AI thinks I say. I was writing to a friend to tell her I looked forward to seeing what her pastries were like and it wrote out " I look forward to seeing what her panties were like".
This guy talks as if we can't differentiate between DeepBach and SebasAnd will anyone actually know the difference?
Or Andrea Bocelli and Luciano PavarottiI wish I shared your optimism. "We" might, but "we are probably in a minority. The majority can't hear the difference between Russell Watson and Pavarotti, or Katherine Jenkins and Renée Fleming.
You are quite right, of course,but nevertheless I believe that is where we are heading, and have already begun.You can't do much to improve the sound when the problem is what was not captured in the recording, i.e. the fundamental problem being the very restricted frequency and dynamic ranges of early recordings. Anything you do requires making assumptions about what might have been lost which, of course, you can never really know. You can try reducing noise but that isn't possible without the high likelihood of also removing some of what you actually want to keep.
But in the case of artists recorded with better equipment there could be a comparison with before and after and some useful adjustments might be made. I am thinking of Ponselle, who had some recordings such as La Vestale that appear to have captured a much closer sound to what she really sounded like and she said so herself.You are quite right, of course,but nevertheless I believe that is where we are heading, and have already begun.
...if you think that the 'discussions' over the validity of the Bruckner 9th and Mahler 10th completions get heated, just wait for thatBut in the case of artists recorded with better equipment there could be a comparison with before and after and some useful adjustments might be made. I am thinking of Ponselle, who had some recordings such as La Vestale that appear to have captured a much closer sound to what she really sounded like and she said so herself.
The questiion is, what was "lost" and what was "diminished". It's a mistake to assume that nothing above a certain frequency (5,000 cps for instance) or below a certain frequency (200 cps for instance) was recorded in, say, an acoustic recording or over long-distance landline transmission (as in certain broadcast recordings). It often is not a matter of hitting a brick wall, or a block filter, but rather of a sharp rolloff--something is there, but not at a very audible level, particularly if there is noise present. This can be amplified to a certain extent, but as of now if you go very far you will be amplifying distortion as well. The trick will be developing software that can discern--and eliminate--what is distortion.You can't do much to improve the sound when the problem is what was not captured in the recording, i.e. the fundamental problem being the very restricted frequency and dynamic ranges of early recordings. Anything you do requires making assumptions about what might have been lost which, of course, you can never really know. You can try reducing noise but that isn't possible without the high likelihood of also removing some of what you actually want to keep.
As far as noise goes, the best thing to do is to find copies of the recording that aren't noisy in the first place. In many cases original stampers exist for commercial 78s and vinyl copies can be made to work from; this has been going on since the earliest LP transfers of 78s. There were also, especially in the US from about 1930-1941, 78s made specifically for electrical pickups from low-noise materials rather than with the abrasives added for acoustic pickups, where the records were made to grind down metal or fiber needles. You could order "Z pressings" (so-called because they have a tiny raised Z in the inner land of the disc, next to the label) from RCA, and Columbia had a line of pressings in a blue material; both play quietly if you can find copies in good shape. This includes custom re-pressing of many acoustic titles.
Noise reduction has gotten very, very good over the past 30 years. Computer logic is very advanced in distinguishing bits of noise from music and filleting the noise out from in between the soundwaves of the music, extrapolating if necessary from the frequencies surrounding the noise. The best, however, still requires close moment-by-moment oversight. It's not enough to just upload a sound file of the whole recording and click a box from a drop-down menu.
1 - 20 of 25 Posts
Classical Music Forum
A forum community dedicated to classical music musicians and enthusiasts. Come join the discussion about arrangements, collections, techniques, instruments, styles, reviews, classifieds, and more!
Full Forum Listing
Top Contributors this Month
View All
Seattleoperafan
491 Replies
Art Rock
453 Replies
Woodduck
302 Replies
Recommended Communities
