AI expert Roberto Nickson successfully recorded vocals with his own voice and used an AI engine to imitate Kanye West’s voice. As a result, the tracks he’s created, including Kid Cudi’s cover of his 2009 hit Day ‘n Nite and his original Kanye West-style instrumental from YouTube, are It was almost seamless and seemed to pretend to be the rapper himself.
“You’re going to hear songs by your favorite artists that you can’t really recognize. I don’t know if it’s them. I’m thinking about all the good and the bad,” Nixon says in the video. I’m talking
The post went viral and received strong reactions on both sides. Grammy-nominated writer and producer Aaron Horne, who recently worked with rapper Doja Cat on her hit Woman, told me: Based on your song, it could be on someone’s record even if you don’t sing the lines.
While this new phenomenon seems like a major breakthrough for music, AI has been in the industry for some time now.
With over 40 years of experience in the music industry, John Burton, Senior Lecturer in Entertainment Engineering at Derby University, has seen advances in AI firsthand. He says the technology has been used in the industry for years as a tool to enhance tracks during the mixing stage of production.
“AI started to get a head start a few years ago, but my first encounter with AI, especially in machine learning, was actually a lot of restoration tools and ways to improve the sound of vocals,” he said. increase. “And you can take an old record with a lot of scratches and hiss and use AI to remove that noise and keep the original vocals.”
Cher’s 1998 hit Believe was one of the first to use groundbreaking auto-tuning software.
/ Getty ImagesAI has brought a breakthrough moment in music. Burton cites Cher’s 1998 number 1 Believe as an example. “It came out a long time ago and became famous because… there were problems with the early software. It wasn’t very good at producing a proper singing voice,” he said.
“Artificially trying to match her voice” is what gave the track its unique metallic flavor. “So Auto-Tune has been around for a long time. Now we’ve gotten really good at it and it’s now built into some of the most popular software. It used to be standalone, very expensive software.” but Apple now sells it [music software] Logic is the company’s standard digital audio workstation. ”
It may have been around for a while, but it has sparked controversy for just as long. Its ability to enhance a singer’s vocals means musicians are divided over its use. Christina Aguilera was famously photographed in 2009 wearing a T-shirt that read “Autotune is for p***ies.” That same year, Autotune helped create her one of the biggest tracks in history, and Ke$ha’s monster hit her single TiK ToK sold her 18 million copies worldwide.
Will Fenton, music insider and owner of music news blog Midder Music, sees a good point. “AI Voice His Changer allows musicians and producers to experiment with different vocal his styles and timbres, potentially opening up new creative possibilities,” he claims. “This technology helps artists discover their own unique sound or emulate the voice qualities of other performers, leading to innovative and diverse music.”
Of course, these changes are also cause for concern, and AI voice changers question the artistic integrity, he says.
“The proliferation of AI voice changers may raise concerns about the authenticity of an artist’s work. Fans and critics alike wonder whether music produced with such technology truly reflects the artist’s ability. Or you might wonder if it’s just the result of AI manipulation.”
This was certainly the case with Ke$ha. Ke$ha eventually got fed up with people claiming he couldn’t actually sing, and in 2011 he created the five-song acoustic EP Deconstructed, stating: To those who said they couldn’t sing. People who don’t like it will hate it, so just do what you want to do.”
AI also poses a tricky situation for copyright law surrounding artistic materials. When computers rather than humans create vocals, it’s no surprise that ownership becomes an issue.
YouTube recently removed Grandayy, a popular content creator with 2.9 million subscribers on the platform, after he posted a video in which he mimicked Eminem’s vocals and made a rap about cats. YouTube removed the video following a request from Universal Music Group, which distributes Eminem’s music, citing the AI-generated track’s similarity to the rapper’s 2009 hit Not Afraid. Issued an infringement removal.
An AI-generated video featuring Eminem’s vocals has been removed from YouTube following a copyright request from his label (Jeremy Deputat/PA).
/ PA media“On the one hand, I completely understand when copyright holders want to protect their art and remove videos that claim or imply they were made by the artist himself, or that imitate original art and compete. I will,” Grandayy said in an interview. “But my video and many others are obvious, fun, transformative parodies that do no harm to the original art. It’s sad to see such videos removed.”
While many of these types of videos are made for fun, it’s clear the industry is nervous about the widespread distribution of this type of content.
The government’s white paper on AI regulation, released on March 29th, claimed an innovative approach to working with AI companies, which is why it drew attention from the creative industry. A review of his Pro-innovation and Regulation, published by Sir Patrick Vallance two weeks ago and prompting the publication of the white paper, originally suggested that: In copyright and access to database materials. “
The proposals presented in this paper would allow AI companies to be exempt from copyright rules, which the Design and Artists Copyright Society (DACS) said could be harmful to creatives. says there is.
George Freeman announced in February that the government would not pursue proposed AI reforms due to concerns from the creative industry (Victoria Jones/PA)
/ PA ArchiveReema Selhi, DACS policy director, is part of the team that produced the comprehensive evidence and briefing that resulted in the government’s failure to proceed with the proposed white paper. We rely on to have some control over the use of our work and to be able to live off of royalties and sustain activity from our work,” she says.
“There are a lot of concerns about your personal data protection or your personal image, or about looking and sounding like you when you don’t. This leads to something that many actors and performers are very concerned about: deepfakes.
In view of these views, the government believes that “the involvement of both the AI and creative sectors will ensure the creation of a balanced and pragmatic code of conduct that allows both sectors to work together and grow.” I believe that,” he said. DACS said it will be part of the engagement process with the Intellectual Property Office once this work begins.
Music Insider Fenton agrees with Selhi’s view. “Using his AI-generated voice, which closely resembles the voice of a real artist, raises legal and ethical issues related to intellectual property, royalties and consent,” he says. “As technology advances, the music industry will need to address these concerns and establish clear guidelines for the ethical use of AI-generated voices.”
While advances in AI are clearly a revenue concern for musicians, Burton notes that over the past few decades, the primary source of income for musicians has fundamentally changed, moving from selling records to live music. points out. It made a big impact.
Sir Patrick Vallance recommends government relax copyright rules for AI companies (Alberto Pezzali/PA)
/ PA Archive“Most performers I’ve worked with have seen big changes. Ever since my career started, you’ve been releasing records to make money and touring to support record sales. I’ve been there,’ he says. “These days, I release records to support tours, because the only place I make money is performing,” he says.
“I’ve seen plenty of examples of Auto-Tune being used live, and it will continue to improve. At this point, it’s going to take years to catch up,” he says, adding that the time it takes to process live vocals delays in technology hindering technological advances in the live performance arena.
Ultimately, the main purpose is to allow artists to work with technology rather than being protected and replaced by it. Of course, the most important perspective on AI in music has to come from the artists themselves. Opinions are divided.
French DJ David Guetta recently created a track he performed with Eminem’s vocals on his February 2023 set, dubbed it “Emin-AI-em” on Twitter. In an interview with the BBC, he described his AI as “music of the future”.
David Guetta says AI is ‘the future of music’
/ PAAustralian singer-songwriter Nick Cave, on the other hand, called AI-powered songwriting attempts “a grotesque mockery of what it means to be human.”
He continued: Data is not affected. ”
Grammy-nominated producer Alex Da Kidd uses IBM Watson, an AI-derived computer system, to create five years’ worth of hits, plus movies, social media, and written articles. Within 48 hours of its release, he reached #4 on iTunes’ Hot Tracks chart.
But if you’re a die-hard Nick Cave fan, you’re unlikely to download it. So with a clearly robust legal framework, wouldn’t there be room for both?
