The Beatles Are Back, Sort Of
-
The Beatles are back (thanks to AI). The first video can be found on Paul McCartney’s YouTube channel. This is a Paul McCartney recording with the addition of John Lennon’s voice as well as rejuvenating Paul’s voice. This was edited by Dae Lims. In the second tune, which was a John Lennon recording, Dae Lims added in the youthful Paul McCartney voice.
The last link is the article that alerted me to this. As mentioned in our ongoing discussion of Bots, the implications are endless. -
What'll they (AI creators) think of next? And do we want to find out?
-
As I understand this, this is just a new voice added digitally. What makes that AI?
-
@kehaulani-0 said in The Beatles Are Back, Sort Of:
As I understand this, this is just a new voice added digitally. What makes that AI?
John Lennon, as you know , died in 1980. Thirty three years later Paul McCartney, in his 2013 voice recorded “New”, the first video in my post. In 2023, Dae Lims created a deep fake of John Lennon’s Voice with AI and added that track to the recording. In addition he either deep faked or modified Paul McCartney’s voice to sound like it did in the Beatle’s Era, again using AI technology.
He also added a deep faked “young” Paul McCartney track to the 1980 posthumous John Lennon release “Grow Old With Me” using AI. Both recordings thus mimic the sound and style of the Beatles using AI technology.
As I said, the implications are endless. For example, imagine Frank Sinatra’s voice putting out contemporary recordings. If this can be done with vocal productions, imagine instrumentalists and others “resurrected”.
We know that AI deep fake video production has been around for a while. Just wait for the 2024 presidential election cycle. If you thought the “Russian Collusion” was convincing, “you ain’t seen nothing yet!”
BTW, is “Kehaulani O” AI or real -
What, pray tell, is "deep fake"? Isn't this just digital overdubbing, or embedding, which has been going on for some time and is a recognised and legitimate technique?
-
@kehaulani-0 said in The Beatles Are Back, Sort Of:
What, pray tell, is "deep fake"? Isn't this just digital overdubbing, or embedding, which has been going on for some time and is a recognised and legitimate technique?
Special Effects and Deep Fake have been going on for a long time. They have become more sophisticated, and with AI technology have been made more usable by ordinary non computer savvy individuals.
The following video from PBS can explain it better than I can: -
@kehaulani-0 said in The Beatles Are Back, Sort Of:
What, pray tell, is "deep fake"? Isn't this just digital overdubbing, or embedding, which has been going on for some time and is a recognised and legitimate technique?
Just go to YouTube and type "deep fake" in the search field; you'll find some interesting results.
-
Got it, thanks.
-
Here is an excerpt from an article about AI Voice Cloning and song generation. The link to the full article is below the excerpt.
What are AI Singing Voice Generators?
An AI singing voice generator is a software application or hardware device that uses artificial intelligence (AI) to generate and modify synthesized singing voices.It is designed to simulate the human process of singing, allowing users to create their own custom vocal performances. The generated voices can be used in various music production settings, such as live performance and recording studios.
AI-generated singing voices can also be incorporated into virtual duets, choruses, and songwriting collaborations. Some AI-based systems are able to learn a user's voice and replicate it accurately, resulting in personalized vocal performances with accurate intonation, natural phrasing, and dynamic expression.
With advances in deep learning technology, some AI-based programs even enable users to create real-time virtual performances that can interact with humans onstage or through audio-visual media.
How do AI Singing Generators Work?
AI singing voice generator technology uses deep learning and text-to-speech (TTS) software to generate realistic singing voices from text. It works by training a neural network on audio recordings of real singers.This process is known as “voice cloning” and it involves capturing the essence of a singer’s vocal range, pronunciation and style in order to produce the same quality of singing that a human would.
The AI then takes the text inputted by the user and converts it into a synthesized version that resembles the original singer's voice. With this technology, users can create their own unique singing voice which they can use for music production, karaoke videos or other applications.
Furthermore, AI singing voice generators are able to analyze an existing piece of vocal audio and automatically generate new versions with different styles or tones without changing the original content. Several companies have already released AI singing products that promise to revolutionize how musicians create vocals tracks for their songs.
Why Have Text-To-Singing Voice Generators Become So Popular?
AI singing voice generators are becoming increasingly popular due to their ability to accurately recreate a vocal performance. With the use of sophisticated algorithms, these tools can take a melody composed by a producer and generate realistic-sounding vocals that can match the original song. This technology has allowed producers to create high-quality vocal performances without having to hire singers or spend time finding samples and creating sample libraries.Additionally, AI singing voice generators offer users more control over their musical projects, as they can tweak parameters such as pitch, vibrato, and inflection to achieve the desired sound. This is especially useful for producers who are on a tight budget and need to create professional-sounding tracks quickly and efficiently.
AI also brings other advantages such as providing new creative possibilities by allowing one singer’s style and tone to be applied to multiple songs in different genres or languages. Additionally, it should also be noted that AI singing voice generators can save time by automatically filling in sections of a song or seamlessly transitioning between two parts of the same track that would have been too difficult for one singer to perform live. All of these factors combined have made AI singing voice generators an attractive option for producers who want quality vocals without having to pay expensive session fees or hire additional vocalists.
https://www.omarimc.com/best-ai-singing-voice-generator-software/
-
@ssmith1226 If that's the best AI can do, we have nothing to worry about.
-
@don-davis said in The Beatles Are Back, Sort Of:
@ssmith1226 If that's the best AI can do, we have nothing to worry about.
That’s what they said about Robert Fulton in the early 19th Century and the Wright Brothers in the early 20th Century.
-
I am waiting for enough AI that we can take this TERRIBLE recording quality and turn it into hifi. The first two examples are so compressed and distorted that I am not sure if it really was AI or just somebody with a similar voice masked by the recording quality. My ears hurt in any case.
Deep fakes are something very real for the arts and I am sure that in 100 years, no one will know what was real or generated.
-
The following, from BBC News, is a story concerning an interview with Paul McCartney on the subject of AI resurrecting the Beatles, pertinent to this discussion.
Sir Paul McCartney says artificial intelligence has enabled a 'final' Beatles song
By Mark Savage - BBC Music Correspondent
Sir Paul McCartney says he has employed artificial intelligence to help create what he calls "the final Beatles record".
He told BBC Radio 4's Today programme the technology had been used to "extricate" John Lennon's voice from an old demo so he could complete the song.
"We just finished it up and it'll be released this year," he explained.
Sir Paul did not name the song, but it is likely to be a 1978 Lennon composition called Now And Then.
It had already been considered as a possible "reunion song" for the Beatles in 1995, as they were compiling their career-spanning Anthology series.
Sir Paul had received the demo a year earlier from Lennon's widow, Yoko Ono. It was one of several songs on a cassette labelled "For Paul" that Lennon had made shortly before his death in 1980.
Lo-fi and embryonic, the tracks were largely recorded onto a boombox as the musician sat at a piano in his New York apartment.
Cleaned up by producer Jeff Lynne, two of those songs - Free As A Bird and Real Love - were completed and released in 1995 and 96, marking the Beatles' first "new" material in 25 years.
The band also attempted to record Now And Then, an apologetic love song that was fairly typical of Lennon's later career, but the session was quickly abandoned.
"It was one day - one afternoon, really - messing with it," Lynne recalled.
"The song had a chorus but is almost totally lacking in verses. We did the backing track, a rough go that we really didn't finish."
Sir Paul later claimed George Harrison refused to work on the song, saying the sound quality of Lennon's vocal was "rubbish".
"It didn't have a very good title, it needed a bit of reworking, but it had a beautiful verse and it had John singing it," he told Q Magazine.
"[But] George didn't like it. The Beatles being a democracy, we didn't do it."
There were also said to have been technical issues with the original recording, which featured a persistent "buzz" from the electricity circuits in Lennon's apartment.
In 2009, a new version of the demo, without the background noise, was released on a bootleg CD. Fans have speculated that this recording may not have been available in 1995, suggesting it was stolen from his apartment, along with other personal effects, after his death.
In the intervening years, Sir Paul has repeatedly talked about his desire to finish the song.
"That one's still lingering around," he told a BBC Four documentary on Jeff Lynne in 2012. "So I'm going to nick in with Jeff and do it. Finish it, one of these days."
'Ropey cassette'
It would seem that technology has now afforded the musician a chance to achieve that goal.
The turning point came with Peter Jackson's Get Back documentary, where dialogue editor Emile de la Rey trained computers to recognise the Beatles' voices and separate them from background noises, and even their own instruments, to create "clean" audio.
The same process allowed Sir Paul to "duet" with Lennon on his recent tour, and for new surround sound mixes of the Beatles' Revolver album to be created last year.
"He [Jackson] was able to extricate John's voice from a ropey little bit of cassette," Sir Paul told Radio 4's Martha Kearney.
"We had John's voice and a piano and he could separate them with AI. They tell the machine, 'That's the voice. This is a guitar. Lose the guitar'.
"So when we came to to make what will be the last Beatles' record, it was a demo that John had [and] we were able to take John's voice and get it pure through this AI.
"Then we can mix the record, as you would normally do. So it gives you some sort of leeway."
However, the musician admitted that other applications of AI gave him cause for concern.
"I'm not on the internet that much [but] people will say to me, 'Oh, yeah, there's a track where John's singing one of my songs', and it's just AI, you know?
"It's kind of scary but exciting, because it's the future. We'll just have to see where that leads."
The star was talking to Radio 4 ahead of the launch of a new book and accompanying photography exhibition at the National Portrait Gallery.
Titled Eyes Of The Storm, the project features portraits taken by Sir Paul on his own camera, between December 1963 and February 1964, as the Beatles were catapulted to global fame.
-
This report from ABC News came out today.