The Ideal Song is not a “Perfect” Song

Without emotion, music is just math. Numbers don't move the soul.

Move me.

Music involves a lot of math: frequencies, rhythms, and repeating patterns that somehow add up to something that works technically. But there's often a key ingredient missing when music comes off an assembly line instead of written with some thought and intent, and then performed with some sense of soul.

There has to be a spark of originality—your own signature sound or vibe.

Perfection is truly the enemy of good music

Yes, a lot of recorded tracks can benefit from some editing but, ironically, it's our human imperfections and our unique personalities baked into the composition and performance that make the "perfect" song. Our ability to both create an emotion and react to an emotion is what makes us human—it's what separates music that makes you feel something and want to hear again from the music that just sounds impersonal, cold and calculated.

Artificial Intelligence (AI) is evolving rapidly with potential real-world application in several fields. Don't assume that creative industries like ours can't or won't be disrupted by these changes too. There are already a handful of automated mastering services that are likely to only improve over time. Tools to identify beats, transcribe melodies, suggest harmonies (and even lyrics) are readily available to musicians everywhere. The key word here is tools—not rivals or replacements for your creativity, but tools to help you create more efficiently.

Efficiency in music-making means making better decisions faster. You get to move as fast as your next idea or instinct, without stopping the creative workflow to complete some non-creative task that could have been delegated or automated while you rock on with your bad self.

How we decide to use these tools is important. If we learn to use our tools effectively and creatively, we can learn to make some great music. If we rely on them too much and allow the tools to take over—or even allow ourselves to become the tools—our music will likely be as formulaic and repetitive as the next line of computer code.

Even if AI is able to compose and perform "real" music tomorrow, it will be limited by the rules already in place. Yes, machine learning is a real thing but, for the time being, it's based in large part on the recognition of existing patterns. As with some of their human counterparts, robot music-makers will ultimately be learning to create more of the same—always staying one step behind and playing catch-up, rather than breaking barriers and moving into new territory.

It will always be the bold artists who were brave enough to try something new who forge the path forward, not the ones who played it safe.

When you become a life-long student of the music you love, you often start by imitating but then move on to creating. Human learning is truly experiential—we learn from the world around us and hopefully adapt to do things differently when we want different results.

So by all means, use the available tools to be inspired and create more music more often. But remember to keep it real by keeping it human. Don’t copy and paste the chorus or other repeating sections—record them each and let the little inconsistencies help keep our attention.

If you want the listener to hear the entire song, be sure to record an entire performance

I don’t mean you have to record the whole song from start to finish in a single take—although that was done for many years. I’m saying that you can’t expect anyone to stay tuned to a 4-minute song when the last two minutes don’t bring anything new to the turntable.

Comping or compiling vocals from multiple takes is something that’s done way more often than many might think, but try to work with an entire phrase rather than a single word or syllable. It will sound more natural and more human, even though it may not be “perfect”.

Don’t always go for perfect. The robots can give us perfect but only you can give us your music and your sound.

I’d love to hear what you’ve been working on. Tell me about your music.

Whenever you find something of value here, please share it on social media and within your musical circles.

Please Subscribe

get thoughts on music-making & making music thoughtful

we respect your privacy