Sent a song

11:12 AM

Sent a song.

I listen to it directly from the messaging app 3, 4, 5 times, because if I download it to my phone I have no idea where it goes.

The mess

Back at the studio, I airdrop it to my computer and slap a single empty vocal track across the top of a new Pro Tools session where I will root around in keyless "mmm's" and "aaah"s, looking for shapes and words that want to find each other, learning the song's emotional trajectory and how it's constructed.

I chunk out the song in blocks, and certain parts come easy.

Dig out the hook first, something of a chant. This will often be the song's title, but sometimes not. Sometimes words need massaged, slanted, or shifted with an emPHASis on the wrong syLLABle to make them work.

Days go by while I continue to finger paint and write in the dirt, chipping away line by line, often over a Mexican lager until I'm ready to type up what I've got. Unfortunately, it seems AI is making this step of the process a lost art. But the messiness is what makes it human.

The rehearsal

Scraps and phrases are dumped into a notebook I pilfered from an old job, while other belligerent parts remain elusive for weeks and sometimes months, until arrangements form in the mist and a scaffold emerges into which enough words will finally fit to be rehearsed.

Rehearsal is a quiet dry run, always done solo and always in whisper to dial in timing and delivery until the song feels calibrated from end to end. Rehearsal refines the lyrics, allowing me to spot the unnecessary, and the things that seem bolted on. The repetition of quiet rehearsal helps me internalize what I'm trying to say, and can last a week or more.

Game day

In the session, tracks are built, color-coded, panned and primed to receive signal. The lyrics I've curated from the notebook are pulled up in a document that lives in the cloud. Headphones go on and the track with the SM7 at the end blinks red.

Topline is double tracked, always. The chorus/chant/hook is a stack of at least 2 doubles on each side - so four or more. Segues and weird changeups are their own thing, so they each get their own track, as do any whispers, talking parts, special effects, drones, throat singing.

These are all carefully mapped with Memory Locations. It can get quite byzantine.

Real-singing begins now, but sometimes it starts days before in the car on the way to the big box store or at home with no one around to prep the throat and diaphragm. Nothing will ruin the start of a session more quickly than raw-dogging a vocal session without warming the pipes first, especially if it's been months since you've sung last.

(I find myself in this position more than I care to admit.)

Takes are recorded late at night when no one else is in the building, and playlists are comped during the day when I can work in headphones. Stacks are built, tuned, bounced, nudged, compressed, effected.

Sometimes I split separation of concerns across different days - knock out the vocal stacks on day one, topline verses day two, bridge and any connectors on day three.

Even with a warmup strategy, my voice is usually thrashed in the days following. Throat Coat helps me bounce back, but regardless, my voice usually sounds like Michael Rooker after a session.

Epilogue

I make a rough mix mockup with throwaway effects, export it to mp3 and send it back over. Someone else is producing and will use the processing they like. If they like it, I'll bounce the takes down into wet and dry folders.

Hopefully not too much time has passed. But in the meantime, the backlog beckons and the work is always there.

++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++

What I’m reading: Faith, Hope and Carnage (Nick Cave and Sean O'Hagan)

What I'm listening to: Animals (Pink Floyd)

Join the Substack: