Paul McCartney is reuniting with the late John Lennon through the utilization of algorithms. Recently, the former Beatles member stated that an old demo recording was reconstructed using AI. It is part of a growing trend of using artificial intelligence to create art, which has attracted both praise and criticism.

"While it is tempting to experiment with new vocal combinations and reimagine classic songs, we must remember that at the core of this technology is the ability to manipulate and distort reality," said Rijul Gupta, CEO of AI company DeepMedia, in an email interview with Lifewire. The consequences of squandering this power are detrimental not only to the individuals involved, but also to the very foundations of trust and truth in our society.

ALSO READ:  Who created Chat GPT – and who owns OpenAI?

Final Beatles Record

AI was used to isolate Lennon's voice from a demo track for the 2021 documentary series "The Beatles: Get Back" directed by Peter Jackson about the recording of the Beatles' 1970 album "Let It Be."

The specifics of how Lennon's voice was revived remain unknown. Robert Harari, the director of the Stevens Institute of Technology's music and technology program, stated in an email that in comparable situations, an algorithm extracts the individual syllables from previously recorded material.

"Therefore, once all samples have been captured and descriptive metadata has been added to the sample file, the next step is to assemble all elements into speech," he explained.

Despite the remarkable technological advancements, Harari believes AI will never be able to imbue a vocal performance with the same emotional context as a human singer.

ALSO READ:  OpenAI Provides $1 Million to Cybersecurity Programs Driven by AI

"Sound production depends on the muscular structure of the throat and the respiratory system," he continued. Significant variation is what makes it sound authentic.

Let It Be?

Digital voices may alter the listener's perception of how the vocalist actually sounded, according to Harari, which is a potential drawback of AI music technology. For instance, the streaming audience is so accustomed to listening to compressed audio that they are surprised when they hear the same recordings on vinyl or CD.

... at the core of this technology lies the power to manipulate and distort reality.

 

"They simply forget or are unaware of the amount of sonic detail lost in data compression for internet media transmission," he added.

Harari noted that there are also ethical and copyright concerns to consider.

"The sound of a singer is their own," Harari stated. "It is their most recognizable characteristic to the ear. It is a misrepresentation when that voice is used for anything other than what the original person wrote, said, sang, or professed. In the current environment, where disinformation is disseminated for any reason, this potent weapon could be used in ways that were never intended."

ALSO READ: ChatGPT enables me to fix code quicker, but at what cost?

Selmer Bringsjord, an AI researcher at Rensselaer Polytechnic Institute, stated in an email to Lifewire that voices are one thing and compositions are an entirely different matter. McCartney producing an AI-written song with Lennon credited as the performer would be a "travesty," according to him.

Bringsjord suggested that in the case of John Lennon, one could contemplate an album about the sudden necessity of violence.

"This would be unethical, but this is the sort of thing that will occur," he added. "This is only the beginning of an avalanche of potential drawbacks. I don't believe Paul McCartney is considering the consequences of his actions. In the context of opera, I believe it would be exceedingly scandalous to "resurrect" a great diva with a "better" voice."

Bringsjord stated that restoring Lennon's intonation is likely just the beginning.

"I'm confident that a new interpreter of the 'American Songbook' will soon emerge; let's call him Sinatra," he added. "Sinatra will purportedly sing better than Sinatra (even when his vast-range tenor voice was still intact), and he will appear in 3D in front of us. That will be extremely eerie."