All These Songs Are Jams!

A beginner’s guide in experimenting with music creation!

As I try to get more composing work, I thought it’d be fun to start with uploading some of my early music experiments and talk about some of the process of creating a jam (please give me a job). If you’re thinking of starting in music creation, hopefully these can give insight into the effects of key, tempo, tone, etc. I won’t go into chord progression or real music theory, just production elements here.

I studied piano for eight years, scored a handful of short films and plays, and think GarageBand is the worst music software ever made but also was complimentary with the 2015 MacBook Air I got for undergrad.

Each experiment here explores how different ingredients of music production impact the final product. I ended up creating dozens and dozens of experiments, but here are a small handful that I think are most beneficial for beginners to learn from.

Pumpkin (Key)

Let’s first listen to a little instruction guide I recorded for my future self on how to play a fun melody I found on the piano. After, we’ll think about how it may translate into a full song on GarageBand.

To me it felt like a song about a sad but regal woman on a dance floor. Hollow Queen was a fun pun on Halloween so I wrote up some spooky lyrics (here’s where I quickly note that my goal is to create music that I myself am not singing, but am writing for others to sing).

This experiment is all about how the key of a song changes how one might experience it. For this, the singer (again hi) has to sing in a lower register, a middle one, and a higher one.

With the lyrics consistent, what do we think this will tell us about the storytelling of a song? The atmosphere of a vocal register? When artists make music, how on earth do they decide the ideal key? The one they sing best in? What if they’re good in many keys? What of our favorite songs might have been even more favorite had the artist just decided it should be sung much lower?

Listen to snippets of these three in any order and decide which ones you think are the most and least successful. It’s okay if you think all are unsuccessful but don’t tell me that.

Again, ignore the singing quality for a second. The lower register in Homemade is uncomfortable, but how does that affect the storytelling? Does it aid with a kind of depressive vibe? Make us feel the singer doesn’t care or that we cringe at the low singing? Personally, I prefer the higher register in Ginger; the singer feels weaker, maybe more subordinate to this so-called Pumpkin Queen.

How does the knowledge of hearing one key affect how we listen to others? What would happen if a musician regularly released singles in multiple keys? Which one would listeners generally pick as their favorite and the “real” one?

Fig (Vocals)

I never listen to lyrics. I don’t know what any of my favorite songs are about. I like music. Let’s take a peek at a little composition with no vocals:

Great. Sometimes when I haven’t written lyrics, I improvise blurbs of words to get a feel for how to make the music around eventual vocals. Let’s ruin this composition with some improv lyrics. I’ll tell you now that I didn’t like the key and lowered the vocals digitally, so they do sound odd. Again, try to ignore that element here.

Great. It’s worse, right? What makes this not work as well? The bad lyrics? The vocal distortion? I felt in general in that this song might do better as a score rather than a song with vocals, and that’s good to learn.

Pear (Atmosphere)

Let’s take some lyrics I wrote in a high school classroom and explore how one might decide already have lyrics and then decide the atmosphere that they want the music to create.

Someone can create a cover of a song that reinvents the thing and makes you ask questions about what the song means — same lyrics, new stylist. If we were to look at the two famed versions of “Both Sides Now” and assess, one recognizes not only how different of songs they are to think about as a listener, but that Mitchell is herself aging and looking back on her life — and that knowledge of the two songs affects how we listen.

What about a song that does a cover of itself right in the middle? How do we assess a song that seems to be repeatedly assessing itself? In Top Chef, they say to always make one good dish rather than divide your time making two, as you’ll be judged on the lesser. Is three songs in one naturally made to fail?

I recorded one version when I had a cold, then every time I got a cold the next year or two, I would record another version for fun. If this is a singer looking back at the same moment over time, how does each iteration of mostly identical lyrics alter the story each time?

I liked the general idea of it, but agree that we as a listener will judge the dish based on the worst iteration, which can often be implicated by personal tastes. I personally like the middle one the best, the third the least, and would skip to the next song when my favorite was done. If I were to create a whole song, I think it may be fun to play with both the first and second iterations; I like the transition.

Another time when sick and interested in creating even more dissonance between the singer and music, I made a fourth version. I like the music but don’t love it with the lyrics, and I think, like Fig, the fourth iteration would be better without any lyrics.

Blackcurrant (Irony)

While we’re still on atmosphere, what happens when a song has nice and sad lyrics and are sung sad — versus sung mischievously and ironically? Do we not pity someone already pitying themselves? Do we have concern about someone not pitying themselves enough given their terrible life? Here’s something I wrote after getting dumped during the pandemic: which version feels better or worse? As a comedy writer, irony is often the name of the game, but does that texture or ruin a song?

Here’s the sadly sung version:

And the ironically sung version:

Do you vibe more with the second person, like they’re not a downer? Or do they harsh your vibe?

Here’s what I think is so interesting about tone: did you even notice that the back half of Blackcurrant is the same music as Pear? Even after hearing mopey versions of Pear in different keys, did your brain feel that Blackcurrant is just another repetition? Or did the different lyrics and intros affect how you understood the music as a whole?

Crazy how you can write a little ditty on a guitar and have a billion different possible songs, right? I love music.

Cranberry (Production)

Now, I like creating songs with low production quality because they 1) lower people’s expectations 2) are so much easier and cheaper to make and 3) give a borderline nostalgic sense of connectivity to the artist. How does the level of digital creation affect how we interact? Here’s a polished song with more production quality (again, not singing quality):

Now let’s hear it but off a recorded piano where the vocals are extremely wet and distant, unclear but still decipherable, and the production elements that are present let us know this isn’t some kind of special session but rather an earnest attempt at a song:

Which song made you feel more connected with the artist and the storytelling? Which one are you more likely to listen to again? If those are different answers, is there something about music where we actually don’t want to connect to the artist? Is the goal for us to appropriate and sing the lyrics ourselves?

As a bonus, I noticed how often slowed versions of popular songs can be all the rage now, so I did a second experiment here with tempo:

Do we again feel disconnected when the singer is distorted? Would it be better if new vocals were recorded at this tempo? Or do we like the slowness and distance? I don’t as much. What would happen if artists released singles at different tempos? Most songs hover around 120 bpm but what do listeners actually want to listen to when given variety?

Kiwi (Embodiment)

Thinking back on how our goal as listeners may be to embody the storyteller rather than listen to them, what about a song where the music is pumping but the vocals are so muddy that you can’t even understand them? Does that frustrate us? Or does it give us room to be the lead singer, this recorded person being our back-up?

Now let’s hear a version of that song but now the music is muddied and the vocals are very present. For the sake of maintaining distance from Emerald’s muddy lyrics, we’ll have new lyrics but still the same hook.

What happens when we can hear the singer? Are we creeped out by him? Seduced? Less connected to the song given this guy is blocking the view? More connected as the singer is who we wanted to hear?

Obviously one song is at a club and one is driving in a car at night — the atmospheres are certainly another variable — but if you created this melody, which song would you want to create with it?

Blueberry (Everything)

With all of this explored, I decided to take the most simple chord progression known to man (key, major fifth, minor sixth, major fourth — Axis of Awesome’s “4 Chords”) and with that create five songs that have the same lyrics but differing atmospheres, keys, ironies, production qualities, ability to hear the vocals, tempos, all of it. Some have small differences, some large. Hell, one even abandons the chords.

Considering that these chords aren’t novel, listen to a random one to start, decide what the song means or becomes, and then see if the others feel like covers or originals.

So, say you’re an artist in the market to buy this song called “And Dream of Tigers” — which one do you want to buy? Remember, you can change it as much as you want, it’s just which one you want to use as a launchpad for your own version. Is it more acoustic? More loud? More sedated? A mix of a few? Any song you write is never really one song.

Summary

Remember that the most option in any music-making software is “Save As” — give yourself freedom to make something new and terrible. I must’ve made twenty Blueberry Jams, and boy are most bad, but some of the better ones are descendants of the rotten.

Here are some songs I made later on, exploring these concepts and making plenty of recipes before I decided I liked these. Please hire me to score your pieces.

Thanks for listening!