Over the last few years, there’s been plenty of speculation over what artificial intelligence could mean for the music industry. Is the modern songwriter a dying breed, or does man have an inherent, creative edge over machines? As most of you probably know, that discourse reached a fever pitch last week with the release of Randy Travis’s new song “Where That Came From.” Employing an artificially constructed vocalist to sing a song is a slippery slope and could lead music down a bland, uninspired path. However, there is somewhat of a silver lining as it gives people like Randy Travis a chance to finally do what they love again when life may otherwise prevent it. It’s a nuanced, futuristic topic that would’ve been hard to comprehend just ten years ago, and plenty of good and bad are worth digging into. Before doing so, though, it is essential to go over the background first.
Background:
On Wednesday, May 1st, Randy Travis shared a video on social media teasing a brand new song. In his message to his fans, the 80s country icon wrote, “You’ll hear much more about the special team of folks who came together to help make this magical moment in my career possible in the coming week. In the meantime, just know that when it comes to me singing songs for you, there’s always more where that came from. Thank you for singing along all these years.”
Naturally, listeners were ecstatic to hear a new song from the legend yet puzzled about how it came to be. Back in 2013, the legend tragically suffered a stroke that took his voice away from him. As it was revealed upon its release, “Where That Came From” was sung by an AI composite of
Travis’s longtime producer, Kyle Lehning, appeared on CBS Sunday Morning to confirm fans’ speculation about how it was made. Lehning revealed they embarked on a creative journey by isolating 42 of Travis’s previous vocal tracks, stripping away the accompanying music, and allowing the artificial intelligence to learn the sound of his voice. They then enlisted the donated voice of Randy’s touring vocalist, James Dupré, as a foundation. From there Travis’s old vocals were seamlessly layered over Dupré’s singing, creating a nearly perfect sound-alike voice that seemed to mirror Randy’s trademark sincerity and soft inflections.
The Good:
The evolution of technology is incredibly divisive. However, one aspect universally agreed upon by all country music fans is that Randy Travis reintroducing himself as a serious mainstream artist is downright remarkable. No one ever thought it would happen again after his stroke, but AI has allowed him to continue his passion despite the circumstances. And by every account, this was something that he wanted to do and wasn’t prodded or persuaded by outside forces. For the first time in the history of popular music, listeners can experience the magic of their favorite legends releasing new music beyond their own time.
Since Travis lost his voice, AI can help him through his disability. It seemed like a distant possibility mere years ago, but today, it’s remarkable how naturally a song like this can come together. Just as individuals with physical disabilities can benefit from prosthetic limbs, those who lose their voice now have access to tools that can help restore their ability to communicate on a whole new level.
It has allowed things to happen that no one thought would ever be possible, even for artists right in the prime of their careers, for an artist without any creative compunctions about allowing a computer to perform their music on their behalf. Zach Bryan has had little time between every one of his releases, and after dropping over 50 songs in just two years, he’s solid proof that fans just don’t get tired of their favorite artists these days. With the help of AI, other artists like Morgan Wallen or Luke Combs, who aren’t as susceptible to quick releases, could theoretically release multiple albums a year while still touring.
The Bad:
Unfortunately, there is a lot more bad than good in the wild, unregulated world of AI. In the hands of artists and record labels who view their output as nothing more than a product to be passively consumed by the masses, introducing such a hands-off approach to music will open a can of worms that could send popular music down a relatively lousy path. Moreover, it may be excellent that AI has given Randy Travis a chance to revitalize his career, but what happens when profiteering labels want to use AI to cash in on a deceased artist’s legacy?
It was recently announced that a new Johnny Cash album will be coming out on June 28th, full of songs written and recorded by Cash before his untimely passing in 2003. That is an exciting way for Johnny Cash to be reintroduced into country music. However, suppose fans are told to accept an artificial version of the Man in Black’s voice as the real thing. In that case, these records are immediately inorganic and potentially damaging to that artist’s legacy. Cash would never allow music not written and performed by him to be released in his name. Not only that but listeners are fed a lie, especially younger ones looking to get into an older artist’s music for the first time. Without the right guardrails in place, misappropriating an artist’s voice without their consent could be commonplace in mainstream music, much like Drake’s use of Tupac Shakur’s voice on a recent record.
While AI could be beneficial because it allows artists to release music more quickly, it remains an inherently dishonest option. What happens to the artistic standards of the music industry if most artists become comfortable allowing an AI to mimic their voice in place of the real thing? It’s a downright lazy way to do business; part of what makes this industry special is the endless work and dedication put into a track. There’s a reason why calling a song “AI-generated” is a popular way to deride boring, uninspired music.
More often than not, listeners fall in love with music because of its authenticity; with AI becoming a viable option, this statement becomes more of an idealistic platitude than a real expectation for the art we listen to. It takes from the entire aura of music, specifically country music, where “real” is one of the biggest selling points. Live sets won’t have the same feel if artists are singing songs that they didn’t have a hand in creating. It strips the integrity of the process and makes it significantly harder to connect with.
Overall:
It wouldn’t be shocking to see legal actions taken to prevent the use of AI, and indeed, that wouldn’t be a bad thing at all. At this critical juncture, it becomes increasingly important for our creatives to be held to a high standard, even if they have no such scruples themselves. However, it won’t prevent dead artists’ families or management from taking advantage of it for whatever reason. Ultimately, this is a case-by-case issue, though it’s fantastic that Randy Travis can utilize it and that he can finally reclaim his voice in the creative process. However, there should be no question that reasonable use cases should be defined. It’s absurd that healthy, contemporary artists should take advantage of AI to avoid the rigor of the recording process. This shouldn’t even be a contentious issue in a world where it’s never been easier to make and distribute music.
