The Good, The Bad & The Incredible of Suno; the First Amazing AI Tool For Music Creation.

I love music. It’s been a passion and a hobby since I was a kid. I play a little, I obsess a lot and I see it live whenever I can. I also love technology, and over the years I’ve tracked when music and technology collide in new and interesting ways, but this is a tipping point. A watershed moment. This is Suno, and this is a rather amazing tool.

Suno is a text to audio generative AI platform that allows you to create music in a way I have never seen before. Anyone with a few ideas and the ability to type them out can create a song in seconds, and for the first time the songs are not that bad. They are crisp, well-produced and sound like some of what you hear on the radio. As always, you get out what you put into it, and this time around what you get out is truly inspiring and interesting with just a little bit of effort.

The basics are Suno has created an AI that trains on segments of audio rather than all the individual pieces of audio it was exposed to so it can learn how to apply these segments and stitch them together in a way that makes sense. I listened to the Rolling Stone podcast where they interview the CEO of Suno and his explanation made a lot of sense, so if you want the details I recommend you listen to him tell it. Suffice it to say that Suno has taken a massive step forward.

Not sure you believe it? Just take a listen:

Here’s a song I created in 10 seconds using “scratching heavy beats trip hop” about ducks: Quack Attack.

And one about beauty in the eye of the beholder to the style of gritty alternative grunge: Beauty in the Shadows.

And lastly, a pop song about the love of a corned beef sandwich: Savory Delights.

Are any of these songs going to be top 10 hits and generate millions of listens? No. Are they able to become the basis for creativity and the ability to generate something interesting, or tell a story? Yes. Suno does some things well to protect artists, like they do not allow you to reference specific artists as foundations for your work. Should an artist be worried about the potential competition from anyone with a keyboard and some extra time? I would argue no. This competitive stimulant is no different than Garage Band or any of the music creation softwares that birthed everything from Nine Inch Nails to Billie Eilish. Talent is still required, and a keen ear will always know the difference.

Music is about the emotional connection created by an artist and an audience. It is a creative outlet that stems from pain, struggle, joy, love or happiness (among other things). What an AI creates based on training data can never compete with the mind of someone who is experiencing love or loss or any of the other strictly human emotions. No machine will ever replicate that spark, and therefore no audience will ever be conned into believing it. A machine may create something catchy, or funny, or interesting, but but it will not be able to create something timeless that resonates for years to come.

All that being said, I see this new AI as inspiring two things. First, it is a tool that any artist can use to create the foundation for a song. I recently heard someone say that AI tools make it possible to tweak endlessly a finished-esque product rather than have to build and rebuild it each time. Allowing an artist to hear the semblance of a finished product so they can know what to tweak, how to tweak it, and then create it in their own styling can be a hugely beneficial way to create. It can speed up the process and allow them to find the right path more quickly than ever before. Secondly, most artists have already come to realize that to make a business from music, you have to connect live. Touring, playing out, and leveraging things like community, fan clubs and merchandise are how you make a living. It is no longer album sales or streaming. Albums and streaming are something you do to create the connection, but monetization of the connection comes from playing live. No AI will ever have a concert to create that connection and no AI will ever create a live, emotional connection with an audience. Simply put, AI-generated music will never have the soul of a live artist.

So should you worry about the tools like Suno? I say no. It should be viewed as a tool and a way to connect with music as a concept. Create and share and use it as inspiration so you can go sit with an instrument and make something more soulful.  

Speaking of great music with soul, check out some of the music I’ve been listening to recently…

Pearl Jam - Running (PJ gettin’ a little punk).

Kacey Musgraves - Cardinal (her best imitation of Laurels Canyon in the 60’s)

The Gaslight Anthem - Ocean Eyes (Billie Eilish cover)

Next
Next

Marketing Philosophy and Sammy Hagar