What is ‘SoundRaw’? The AI tool music embraced in 2024
Over the last couple of years, the discussion surrounding the moral ambiguity of using artificial intelligence in music has risen significantly, and with AI tools being developed at an alarming rate, it’s likely that the conversation will only become more heated the more prevalent it becomes. There are many artists who have spoken out against the use of AI in music in recent months and are rightfully concerned about the way in which it diminishes the hard work and effort put into the creative process by making it so simplistic and detached from the very human experience that has traditionally been a part of songwriting.
With that in mind, why have some artists been embracing tools such as SoundRaw, the AI beatmaking and music generating automaton that has taken off in 2024, and how is it dramatically affecting music creation in a way that purports to be “the future of music production”?
Let’s get one thing straight – AI has not completely destroyed music, and until it develops in a way where human existence becomes entirely obsolete, then it won’t destroy music. There will always be a demand for authentically produced music in all genres, and the consumer will argue that nothing beats something that was created with a human touch. However, the strength that tools like SoundRaw possess in the ability to create admittedly rudimentary electronic beats and try to mimic the voices of artists without their permission is what is most disturbing and poses the question of whether some people who lack the ability to challenge what is real and what isn’t will become complicit in AI’s rise in popularity over the coming years.
On their website, SoundRaw states that they have trained their machines in an ethical way, claiming that “everything our AI learns from is produced in-house by our dedicated music production team” and that it doesn’t use copyrighted material to generate beats and sounds for producers to use. At the same time, they appear to take petulant swipes at other similar platforms such as Suno AI, for how they “tend to lack the depth and individuality that platforms like SoundRaw achieve through close artist collaboration and customisation”.
In their recent countdown of the top eight AI-generated songs that were produced in 2024, not all of them were produced using their own platform, with a remix of The Weeknd’s ‘Blinding Lights’ featuring the vocals of deceased artists such as Michael Jackson and Freddie Mercury being used in less than tasteful or believable fashion being included on the list alongside some SoundRaw ‘originals’. The trouble is that the originals that the tech company claims to have had a hand in producing, such as French Montana’s ‘Ride the Wave’ and Trippie Redd’s ‘#RBDG’ are so devoid of any personality, creativity or originality, that it instantly depreciates all of their claims that they have more depth than other AI music generators.
If you’re content with the music you listen to lacking any substance and simply sounding like a half-assed attempt to sound vaguely similar to other banal hits you’ve heard in the past, then sure, the commodification of music production in a way that’s accessible to all via the means of AI really is a fascinating development in technology that deserves praise. However, in a world where crypto scams and pyramid schemes seem to also entice people if you’re also falling for companies telling you that AI is fully capable of making interesting music or that you can call yourself a producer if you’re using AI tools to make your music, then bad news – you’ve been duped.
There’s an argument to be made that those who have programmed their own artificial intelligence robots, such as SoundRaw, are musical creators since they are feeding the technology with everything it needs to learn rhythmic patterns, create melodies and understand harmony, but if you’re simply using a pre-existing tool and claiming it as your own, then that’s a cop-out. If anyone wants to make the argument that hip-hop and electronic music have been doing this for many years with the use of presets and sampling, then that’s true to some degree, but they’re still being manipulated in ways that involve human involvement and aren’t simply reliant on a computer to spit out the first piss-poor loop it can come up with.
There’s only one way to ethically use AI in music or art, and that’s for you to actually contribute towards the machine learning process in the first place, in a similar fashion to how Holly Herndon achieved this on her 2019 album Proto, where she taught an AI to synthesise her own voice and essentially made herself into an instrument. Falling for a scam that claims it can elevate your production skills to the next level by churning out musical detritus isn’t ethical. It’s moronic.
