But what if you throw other stuff at PaulStretch? It turns out it does a pretty nice job of turning anything into a variation on gooey ambiance, though the results do vary in the details from song to song.

So here’s the question: can you recognize super-slow, super-smoothed versions of popular songs? I’ve picked two dozen highly recognizable songs at random out of my music library, fed small you’d-know-it-if-you-heard-it clips of each into PaulStretch, and presented the results below. Give ‘em a listen, take your best guesses in the comments if you like.

1-6
Download audio file (01.mp3)

Download audio file (02.mp3)
Download audio file (03.mp3)
[Oops! #4 was a dupe! -j]
Download audio file (05.mp3)
Download audio file (06.mp3)

7-12
Download audio file (07.mp3)
Download audio file (08.mp3)
Download audio file (09.mp3)
Download audio file (10.mp3)
Download audio file (11.mp3)
Download audio file (12.mp3)

13-18
Download audio file (13.mp3)
Download audio file (14.mp3)
Download audio file (15.mp3)
Download audio file (16.mp3)
Download audio file (17.mp3)
Download audio file (18.mp3)

19-24
Download audio file (19.mp3)
Download audio file (20.mp3)
Download audio file (21.mp3)
Download audio file (22.mp3)
Download audio file (23.mp3)
Download audio file (24.mp3)

For the acoustics/processing nerds: the software does some fairly aggressive audio processing to manage to create slowed-down music that doesn’t sound like choppy hell—as a result, it leaves a fairly strong acoustic imprint on everything it touches, and between the major tempo shift things get really smeared. I threw System of a Down’s Chop Suey at it and the rapid-fire vocals just turned into big echoey smears.

This is all tweakable, of course, and the Windows version at least is incredibly simple to get going. I highly encourage you to give it a shot if you’re at all interested in this output. Just download, extract, and start feeding it audio files and messing with the controls. I kept the settings at default for all of these: about 8x expansion in time, about 7k samples. Play with both, you may find that different settings work particularly well for different songs.

This is all of them.

If you know of any not featured here, or related folk-musical artifacts or oddities tied to the subject, leave me a comment and I’ll be sure to get them added. If you’re curious about the backstory to some of this, see my previous post about the Slater folks music / viral weirdness thing.

The various songs of Steven Slater

My version, lyrics by Max Sparber:

Rachel‘s Ballad of Steven Slater

Jonathan Mann‘s Ballad of Steven Slater

Ornaith O’Dowd‘s The Ballad of Steven Slater

Gary Morton‘s The Steven Slater Song

DanReynoldsRock‘s Steven Slater Jet Blue Blues

Gary DeLena‘s 50 Ways To Leave Your Dayjob

Andy Hoskinson‘s Steven Slater Lost His Cool

Noah Needleman and Lee Curreri‘s Jet Blues

Steve Kolander‘s Fly Away

Elliot Pollack‘s Current Events Song #1: Steven Slater

JudyCoconutty‘s An Ode to Flight Attendant Steven Slater of Jet Blue

rappingFA‘s Steven Slater Resignation Rap

hidingtruid‘s The Ballad of Steven Slater

Midstokke‘s The Ballad of Steven Slater

ondancetronetarnatha‘s The Ballad of Steven Slater (after John Wesley Harding)

AirPurifierReview‘s misleadingly-titled Take This Job and Slide It

UpsetMama‘s Goodbye Now, Take Care, I’m Off!

GuitarConi2‘s Steven Slater Song

Mighty Bulger‘s Steven Slater Song

Late night talk show host Jimmy Fallon performed his (or his writing staff’s) version three times last week:

8/10

8/11

8/13

BrooklynBlowback‘s The Ballad of Steven Slater:

Related oddities

My interest has largely been in collecting actual recorded musical compositions/performances, but there are few things I’ve tripped across that fall more on the periphery of this collection than anything. Again, let me know if I’m missing anything interesting.

Here’s a cover of Rachel’s ballad by youtuber Kimfob:

IGN blogging-type-person That Reilly Monster wrote lyrics to a ballad

CartoonOverlord put together an odd little animation complete with jingle, Steven Slater – Fabulous Quitter

Blogger Adam Kovac does a conspicuously grumpy roundup of (some of) the Slater songs.

This is probably the only time I’ll ever end up being the soundtrack to Wolf Blitzer’s “The Situation Room”.

So: I recorded a song yesterday morning, using the lyrics of fellow Metafilter user Max ‘Astro Zombie’ Sparber. The song is called Ballad of Steven Slater. It’s a little three-minute pop-rock tune that I’m pretty fond of, especially for something I knocked out in only a few hours. Here, give it a listen!

Download audio file (2010_ballad-of-steven-slater.mp3)

UPDATE 8/14: now with music video! Check it:

Of course, you might already have heard a bit of it yesterday, if you happened to be watching CNN…

Or, if you’re a Portlander, you might have heard a bit more of it on KGW TV’s “Live @ 7” broadcast, where they played the whole bridge with lyrics on the screen—and understandably bleeped and blurred the enthusiastic “fuck” at the capstone of same:

Or, on Good Morning America today, you might have caught a recitation of the end of the first verse at the end of this clip.

I didn’t do anything but record the song and post it over at Metafilter Music, which is the sort of thing I do on a regular basis. Usually Wolf Blitzer doesn’t end up involved. So what the hell happened there?

Meming the Cube

Steven Slater’s sudden showy departure from his career as a JetBlue flight attendant has been making the rounds since the news broke Monday morning; by Tuesday morning it had outlived the typical too-good-to-fact-check news-of-the-weird wire blurb’s lifespan and turned into a genuine meme. The dominating feature of that meme: Slater as folk hero.

Here’s a man doing what we all wish we could do, being who we all wish we could be. A man who, having had enough, says: fuck this noise, me and my beer are taking the airplane slide outta this bullshit. If anybody needs me, I’ll be fucking my boyfriend.

And folk heroes get folks songs, right? The idea is obvious enough that people were suggesting the idea in idle tweets yesterday morning. Seattlite Caleb Hannen, meanwhile, suggested that Slater isn’t a folk hero because he doesn’t have a folk song, right around the time that I was posting mine to the internet. Heh. (Take heart, Caleb: Arsenio Hall isn’t a fan either.)

But before hellocruelworld tweeted or Caleb blogged, Max Sparber had already written up six stanzas about Slater and posted them to Metafilter. And when I came back from breakfast at my local diner, I saw the lyrics and got to thinking, and after some noodling on the piano decided I had a song worth recording.

So I spent the next three hours arranging and recording the ballad, and I posted it to Metafilter, and people seemed to get a kick out of it. Mentioned it on twitter and a few people passed it around, which is always nice to have happen. Success, as far as I’m concerned; I kept half an eye on my Tweetdeck notifications and went back to my dayjob moderating MeFi.

But then it kept making the rounds. Waxy linked to it. Gruber over at Daring Fireball linked to it. One of my favorite people, mefite ColdChef, nudged fellow mefite and CNN video editor type person Vidiot, who managed to get it on to the Jeanne Moos segment above. A local news channel picked up the song without even realizing initially that I was a Portlander.

Max, savvy fellow, let his local press know about the whole thing, and so I got mentioned variously in a few different pieces as “songsmith and MetaFilter moderator Josh Millard“, “one of the dudes that helps run the site“, “some other random guy“, and “another musician“. If I ever get to Minnesota, I will exploit this to the hilt.

So it’s been surreal. Fantastic. I couldn’t have planned it. If I’d known the song would get so much (granted: random, faddish, fleeting) exposure, I might have spent more time polishing up the vocals. But if I’d taken more time, I might not have gotten it out promptly enough to let all of the above happen.

The Competition

Timing probably mattered. My collaboration with Sparber was certainly the first Ballad of Steven Slater to really hit the scene with force, but folk music is by definition a populist phenomenon: if Max could write a ballad, so could other people. And so there were other takes on The Ballad of Steven Slater (with, reliable trope that “The Ballad of ___ ___” is, precisely that name) popping up yesterday.

Youtube user thestratbrat posted her very-folky take on Steven:

Daily song machine Jonathan Mann went likewise acoustic-folky, and adapted Slater’s verbatim intercom blue streak as his chorus:

Even Jimmy Fallon, later that night, milked the idea, though he gets pretty rapidly distracted from actual Slater-related balladeering in favor of Sandler-alike call-and-response jokes:

And much as I was riding high yesterday on the link love and the twittering and the blips of TV footage, I can’t compete with NBC; Fallon’s bit is currently dominating the twittersphere. Alas. I’ll always have CNN. Though if any national talk show producers feel like having me on for a live performance, I’m pretty sure I could fit it in my schedule.

I’d like to think there’ll be more takes on Slater coming along, but news of the weird tends to die fast and memes don’t live much longer. Four songs about one guy’s bad day is pretty decent, in any case, and it’s been fun if weird to be caught up so directly in the whole zeitgeist like this.

Singing The News

This isn’t the first time I’ve worked with Max Sparber (or, to better characterize the situation, taken his lyrics and run with them—he’s in Minnesota, we’ve never met, and we’ve never intentionally coordinated this stuff) on news-related stuff: a couple years ago, I recorded his epic ode to a Slidell kid who managed lose his arm to an alligator but come out on top, in a tune called Young Devin Funck:

Download audio file (2008-josh-millard–young-devin-funck.mp3)

And late last year I recorded fellow Metafilter member mikepop‘s lyrical adaptation of Darling Corey to the hoax spectacle of the ostensibly aeronautical Heene family, Oh Balloon Boy:

Download audio file (2009-josh-millard–oh-balloon-boy.mp3)

Beyond that, though, I spent fully half of 2006 writing short songs several times a week about developing news stories, in a now-long-defunct project called The Aural Times. It was a fun but stressful project; after a few months I burnt out and my output dropped to a crawl and then to nothing. (I had begun to redesign the site as a general music blog, before abandoning that idea as well and eventually building this site from scratch instead as my general musical archive; accordingly, much of the hokey faux-newsprint charm of the original is currently swapped out for a more slick and neutral layout, something I really ought to fix at some point.)

I recorded something like seventy songs during those six months; you can find them all in the Aural Times archives, complete with writeups from my Editor-In-Chief doppelgaanger; see for example the July 2006 material. But I’ll link a few of my favorite recordings here for quick reference:

Download audio file (america-online-releases-twenty-million-private-search-records.mp3)
Download audio file (bush-leaves-for-india-pakistan-tour.mp3)
Download audio file (dan-rather-leaving-cbs.mp3)
Download audio file (german-cat-had-deadly-strain-of-bird-flu.mp3)
Download audio file (homer-eagles-will-be-beggars-no-more.mp3)
Download audio file (paraplegic-woman-runs-from-cops.mp3)
Download audio file (sleater-kinney-calls-it-quits.mp3)
Download audio file (woman-attacks-dog-breeder-with-body-of-dead-chihuahua.mp3)

Serial project-starter and project-abandoner that I am, The Aural Times is easily the project I miss the most, and that I most wish I could have found the time and energy to keep up. While as an official project it may be dead and buried, things like yesterday’s Ballad of Steven Slater recording are a nice reminder of what I enjoyed about the project, and for the handful of stray Aural Times fans still out there on the web I hope these occasional forays into longer-form news-related projects are a bit of a consolation.

Endnote

While my recording was making its way onto CNN yesterday, somebody went and ousted me as mayor of the diner I’d just had breakfast at. Makes me so mad I could hit something. Probably a piano. With my fingers. Rhythmically. Hmm…

Misc. further linkage

- Margaret Lyons at popwatch.ew.com gives Max and I a nice nod.

Last night, Jimmy Fallon sang “The Ballad of Steven Slater.” And it was pretty good. Except I’d already heard a different song called “The Ballad of Steven Slater,” and that one is mas macho. Or something. Take a listen (contains salty language)…

This one, with lyrics by Max Sparber and music by Josh Millard, packs a certain poetic punch I feel the Fallon version lacks…

- NYMag, Daily Intel, Steven Slater: the ballads:

But the best tunes so far are by Jimmy Fallon, and MetaFilter’s cortex and Astro Zombie. “It ain’t that easy to ride the skies / laboring for JetBlue / A man’s got to keep widened eyes / For terrorists or shampoo,” goes the latter. “And worser still are the passengers / They turn a kind man to a hater / Won’t nobody stand up to this? / One man: Steven Slater.” (Seriously, go listen to it.)

- Sundance channel, SUNfiltered:

Over at Metafilter, one MeFite Astro Zombie penned “The Ballad of Steven Slater” which Josh Millard aka Cortex (a MeFi moderator) then put to song. It’s fantastic.

- Yet another ballad, this one an angular poprock composition from Irish Brooklynite Ornaith O’Dowd.

- Fellow balladeer thestratbrat managed to show up on a CNN segment as well, today, and on a WPIX NY segment (about 2:00 in on each). Judging by the comments on her Youtube post, she’s even more surprised to end up on CNN than I was.

- Yet another ballad, though only lyrics this time, from blogger That Reilly Monster.

- And for god’s sake, Fallon, a reprise?

- Jonathan Mann’s version gets used as background music for an ABC Australia radio piece.

- Sparber’s lyrics excerpted in a Philly Inquirer writeup about the Slater/Folk Hero thing.

- Dutch Radio 1′s “BNN Today” did a little Segment on Slater ballads, featuring excerpts of mine and Max’s, Rachel’s, Jonathan’s, and even the accursed Fallon’s.

- Youtube user Kimfob recorded a cover of Rachel’s version. 2nd generation folk song! (Linked video is a keychanged do-over; he appears to have removed the original.)

- Jonathan’s version gets blipped briefly in…a sort of spammish-seeming market-watch talking head type thingy. Okay!

8/12: Some more aggressive searching turned up a number of other Slater songs I hadn’t seen yet (or which hadn’t yet been posted)!

- Greg Morton with a new Slater tune, a jaunty Welkian short that breaks with naming tradition in favor of alliteration: The Steven Slater Song

- Youtuber (duo?) DanReynoldsRock with the, I dunno, yacht-pop? Steven Slater Jet Blue Blues

- Beret-wearing Gary DeLena with a Paul Simon parody, 50 Ways to Leave Your Day Job

- Andy Hoskinson with a fairly critical take: Steven Slater Lost His Cool

- Noah Needleman and Lee Curreri cover the White Guy Blues angle with a solid guitar-and-piano tune, Jet Blues

- Steve Kolander’s breezy stroller, Fly Away

- Elliot Pollack looks like he’s kicking off a project after my own heart with his Current Events Song #1: Steven Slater

- And, filling a key niche, here’s retired flight attendant JudyCoconutty with her rap composition An Ode to Flight Attendant Steven Slater of Jet Blue

- In the same vein but with a skosh more production value, here’s rappingFA with Steven Slater Resignation Rap

- In a bit of opportunist re-branding, band Mind’s I retroactively declares their existing track Jaded (As A Flight Attendant) as an homage to Slater, with a thematic video to back it up.

- A weird brief animated joke promo, more jingle than song: Steven Slater – Fabulous Quitter

- Here, from hidingtruid, is another Ballad, sounding a lot like the verse to Happy Xmas (War is Over).

8/14: Another bit of searching, another crop of songs:

- Seattle band Midstokke short shouty The Ballad of Steven Slater

- ondancetronetarnatha apologizes to Dylan for his folky The Ballad of Steven Slater

- Jimmy Fallon does it a third time. I swear to god.

- In CNN news (geddit), this schizophrenic montage of web reactions includes a fair bit of both Jonathan and Rachel’s videos.

- On the metacommentary front, Adam Kovac doesn’t like these songs one bit.

- And here’s a ultra cheapo karaoke-assisted parody of The Box Tops’ classic tune “The Letter”, with the misleading title Take This Job and Slide It, from weirdly shillish youtube account. Weirdness. (But considering how many people have been just tagging their random old videos with “Steven Slater”, this counts as making an effort by comparison.)

- GuitarConi2 knocks out Steven Slater Song from her easy chair.

- upsetmama put together a waltzy take called Goodbye Now, Take Care, I’m Off

- Mighty Bulger does some minimalist poprock stuff in Steven Slater Song

Could I easily and algorithmically make recordings more monotone?

The answer turns out to be “sorta”. I’ve written a simple python script that uses the Remix library to do some fine-grained pitch-shifting to the songs I throw at it, and the results vary widely from song to song for a variety of reasons I’ll explain below with some examples. If you want to skip the nerd talk and jump straight to the weirded-up music, see the bottom of the post for a collection of songs I ran through the script.

Boringization

Here’s an example of a song that produces very clear results: Pink Floyd, Comfortably Numb

Download audio file (tcomfortablynumb.mp3)

Note the that song seems to pretty much hang on a single chord throughout (or more specifically seems to idle back and forth between a major and a minor version of the same chord), while the melody and the guitar solos seems to hop around into weird jarring normal-then-chipmunks-then-normal territory. I’ll talk in more detail about this in a bit, but the key thing happening here is that the script is shifting basically every chord in the song to the tonic, the root key of the song. So instead of a chord sequence like Bm -> A -> G -> Bm -> D we end up hearing Bm -> B -> B -> Bm -> B.

Note also just how flat the song becomes in the process. Instead of big satisfying chord changes and nice minor-into-major catharsis when the verse breaks into the chorus, it’s all just…samey. This is actually something like an ideal outcome for the monotoning process, if not an ideal way to produce engaging recordings.

My friend Roj suggested a good alternate name for this process: “boringization”. When the process works well, it strips out most of the musical movement in a song, producing something far more static, at best sort of hypnotic and at worst simple dull and uninvolving.

One Note Pony

Here’s a far simpler example: Suzanne Vega, Tom’s Diner

Download audio file (ttomsdiner.mp3)

With an a capella track, the job of the script is very simple: figure out what note the singer is singing, and pitch shift that to whatever the tonic of the song is supposed to be. The result is Suzanne Vega singing one note over and over again. (Or nearly; the process’s pitch detection is very simple and imperfect and so it can for various reasons be fooled a bit, so we get moments where she’s shifted to the wrong note.)

And while the note remains the same, the quality of her voice changes significantly from note to note. This is a result of the pitch-shifting being done—when you take a complex sound like the human voice and shift all the frequencies up or down significantly, the result will generally not sound right because the distribution of frequencies in your pitch-shifted sample is different from the frequencies present in that person’s voice when they’re actually singing that lower or higher note. In other words, the timbre, roughly the frequency qualities of the voice, are changed.

This is why voices sound funny when music is played back at the wrong speed on an analog music player (e.g cassette deck or vinyl turntable), hence the Alvin and the Chipmunks or Boomy Satan effects everybody is familiar with regardless of whether they’re an acoustics nerd. (The effect of inhaled helium or xenon on the human voice is not unrelated—the differing density of those gases induces a timbral change in sound created while it’s being exhaled through your larynx. But that’s a different discussion.)

Desperately Seeking Tonic

So what is actually going on with the monotoning process?

Every song, before my own code does anything at all, gets put through a process (the Echo Nest “Analyze” function) that looks in close detail at the composition of a recording. That process does a lot of things, but only three are particularly important for what my script is trying to accomplish. Those three things are:

1. Cutting the song up into very small pieces called segments, usually some fraction of a second long each.
2. Analyzing the frequency content of each of those segments to produce a list of twelve pitch values that correspond to the twelve notes on the familiar western chromatic scale (e.g. the twelve black and white keys that make up an octave on a piano keyboard, from A to G# by half-steps or semitones), with a larger number for any given pitch value meaning that more of the frequencies that correspond to that note are present.
3. Analyzing the whole song to make a guess what its tonic note is (that is, what key is it in, e.g. C or F or G# minor).

What my script does is take that tonic note from (3) and treat that as the note that we want to rule over the whole song. Then, for every single little segment of the song, in order, the script looks through that segment’s array of twelve pitch values and finds the one that’s the biggest — roughly, the note on the scale that is loudest for that little bit of the song. And it compares that note to the tonic note. If it’s they’re the same, nothing happens.

If the tonic and the detected pitch are different, the current segment gets pitch shifted by a number of semitones equal to the distance between the detected pitch and the tonic.

So, coming back to Comfortably Numb as an example, if the detected tonic of our song is Bm (the first chord of the song and the one it keeps coming back to), then it won’t shift the the first bit of the verse (“Hello / Is there anybody”) at all; but when the verse changes from a Bm chord to an A chord (“In there / just nod if you can”), the script notices that instead of a big fat B being played by the bass, it’s now a big fat A. A is two semitones away from B (A, A#, B), and so the script shifts those segments in the A chord portion of the verse up two steps. Et voila, instead of the chords going Bm -> A we end up with Bm -> B.

Likewise, with the Tom’s Diner example we saw (pretty much) every single note of the song shifted to the tonic note.

Fighting For Attention

This is complicated, though: with the exception of very simple cases like a solo a capella performance, songs tend to have a lot of notes happening at any given time. You’ll usually have, at a bare minimum, some major or minor chord triad playing on keyboards or guitars or other polyphonic rhythm instruments, a bass line playing something that may or may not stick to the root of the current chord, and a vocal melody that may (in fact usually will) stray constantly from the root of the current chord. Mix in stuff like lead guitar and vocal harmonies and things can get very sonically complicated.

So how does the script deal with this?

It doesn’t, in any clever way. It’s about as stupid a script as you could hope for; it does it’s one job (“Fetch the loudest pitch of the current segment, boy! Attascript! Who’s a good script! Who’s a gooood script!”) and doesn’t care what part of the mix is responsible for making that dominating bit of noise.

Which means that what happens to a song when it goes through the monotoning process depends largely on how that song was recorded and mixed in the first place.

There are three general outcomes that I’ve seen in my initial experiments:

1. Dominating bass/rhythm mix. Comfortably Numb is a great example of this: the bassline is steady and the rhythm guitar and strings playing the chords are as well, and they’re mixed loud enough that in any given segment they tend to be producing a big fat tonic note that keeps the chords really recognizable. The script ends up keeping the whole song very single-chord as a result, and the vocals sort of jump all over the place as they get shifted around.

2. Dominating vocal line. Tom’s Diner is a trivial example of this, since it’s all vocal, but any song where the vox are mixed relatively loudly compared to the rest of the instrumentation there’s a good chance you’ll see a similar effect: the vocals stay on pretty much one note (or two notes, an octave apart, jumping back and forth, for stupid-pitch-shift-calculation-algorithm reasons), and the rest of the song will flail around in the background with very weird chord changes.

A good example of that happening in a full mix is the Crash Test Dummies‘ mid-90s earworm Mmm Mmm Mmm Mmm, in which Brad Robert’s seismic lead vocals are present enough in the mix to catch the script’s attention most of the time, yielding a very weird and far less catchy rendition in which he seems to be trying to destroy some architecture through meditative humming:

Download audio file (tmmm.mp3)

3. Total madness. If neither (1) or (2) holds, that generally means there’s no one characteristically dominating sonic element in the mix, and so the script ends up following different pieces of the mix from moment to moment. Here the bassline, but then the lead vocal, but then the solo guitar line, but then the synthy thing that swelled up a bit, and so on. Songs along these lines tend to be pure weird chaos to listen to, which is great if you’re into that sort of thing (hi!) but makes for less compellingly clear “look at this specific thing that’s happening” examples.

See for example The Shangri-las, Leader of the Pack, which goes from charming Spector-era pop tragedy to horrible horrible nightmare music:

Download audio file (tleaderof.mp3)

Of the several dozen songs I tried, the collection I’ve posted below mostly fall toward types 1 and 2, though there’s a couple that can’t seem to make up their mind and so are either sort of 1/2 hybrids that go back and forth or outright type 3 messes.

These Are A Few Of My Favorite Monotothings

Download audio file (tamerican.mp3)
Download audio file (tblister.mp3)
(love the monotone guitar riff)
Download audio file (tcakesurvive.mp3)
Download audio file (tconstantsorrow.mp3)
(monotone bluegrass is actually plausible)
Download audio file (tfrienddevil.mp3)
Download audio file (tigothigh.mp3)
Download audio file (tinthegarage.mp3)
Download audio file (tistanbul.mp3)
Download audio file (tkarmapolice.mp3)
Download audio file (tmeandagun.mp3)
(another a capella example)
Download audio file (tpennylane.mp3)
Download audio file (tringoffire.mp3)
Download audio file (tsanteria.mp3)
Download audio file (tsaveyoursoul.mp3)
Download audio file (tsgtpepper.mp3)
Download audio file (tsomebodytolove.mp3)
Download audio file (tsympathy.mp3)
(classic rock = type 1?)

Notes For Improvement

I’ve got a few ideas in mind for how to improve/expand this script.

One obvious problem with it is that it’s very twitchy: even if when doing a pretty good job of creating a steady monotone feel (whether vocally or in terms of bass/chord lines), it’ll have little moments where it freaks out because of some transient element in the music and leap to a way different note for half a second. If I were to create a smoothing function that looked for areas that were steady and ignored brief variations between those steadier bits, that’d probably help along with glitch reduction.

Another issue, harder to deal with automatically, is controlling the balance of pitchshifting up vs. pitchshifting down. A couple things come into this:

First, some songs have the melody mostly move around in the notes between the tonic and the fifth of the scale, whereas other songs have a melody that dives down below the tonic a bit. Choosing to shift up or down is the difference between Satan Voice and Chipmunk Voice, and bigger shifts have more blatant effects. So it’d be good to try and pick a balance point where neither the up nor the down of the vocal is any larger than it has to be. Striking that balance point would depend on analyzing the melody, which is pretty easy to do with a human brain and a quick listen but is harder to do with my narrow existing programmatic toolset.

Secondly the pitch info available for a segment is only a list of twelve notes on an abstract scale — you can find out what named notes are present, essentially, C and F# and A, but you don’t know what octaves those are in. Which means that knowing whether the current note is “below” or “above” another detected note in a different segment is impossible at a glance, which can lead to situations where instead of ideally pitching G4 (i.e. a G note in the fourth octave) up two semitones to A4 and B4 down to semitones to A4 (with minimal timbral whackery), the script can end up shifting G4 down ten semitones to A3 (hello, Satan!) or B4 up ten semitones to A5 (Alvin, Theodore, et al).

There may be good ways to deal with that stuff more elegantly. I haven’t tried yet.

I also wonder if I can use the segment timbre information that Echo Nest provides to make some sort of educated guesses about what kind of sound or sounds are responsible for generating the current dominating pitch. It might be possible to at least partially differentiate between e.g. round bass notes vs. sharper vocal or guitar notes and do some intentional targeting of one part of the recorded mix vs. others for what the script pays attention to.

Another possible approach to this whole idea would be to look not just at the single loudest pitch for every segment but at all the pitches present, and try to identify chords based on the three or four most present pitches in the segment. If you’ve got proportionally a lot of C, E, and G standing out in a segment, it’s a good guess that you’re dealing with a C major chord, whatever else might be going on. Combining this sort of chord detection with the glitch-smoothing idea above might make it possible to analytically ignore parts of the mix and really just focus on the chords underlying a song (to pull off a stronger type-1 It’s All One Chord effect) or to attend to the vocal line when it seems to stray from the chord roots (to better achieve a type-2 The Singer Only Knows One Note thing).

That last idea might also make for a way to build a rough chord-analyzer system, which would be a boon to anybody who needed the chords for an mp3 and couldn’t find nice simple tabs for it anywhere.

Anyway, I’ve had a lot of fun with this so far. If you have specific ideas for how to approve or differently approach this idea, or have a song in mind that you think might producing interesting monotonization output, let me know.

The new album features a lot of interesting live-looping work that Hiram has been doing, and it’s good hypnotic fun. (I’ve had the pleasure of seeing him do some of this stuff live solo as well, which is impressive in its own right as a feat as much technical/choreographical as it is musical.)

While you’re at it, go listen to this great cover of Spanish Bombs that The Harvey Girls did in collaboration with the fantastic Alantl Molina and Verónica Bagnoli.

Here’s the music track in isolation (or download the mp3 here):

Download audio file (josh-millard–nine-inch-niles–the-crane-that-feeds.mp3)

But, a little background on what’s going on!

So it’s been interesting seeing reactions to The Seattleward Spiral: some folks love the concept and hate the execution, some folks like the raw weird noise of the thing, and a lot of people have opinions on how it might work better. And certainly there are a lot of different things that could in theory be done to make it work better — if I had known the project was going to get so much attention I would have taken it a little more seriously in the first place. Heh.

But, so! How to do it better? There are a lot of possibilities; unfortunately, many of them involve rewriting portions of afromb.py, which I’m frankly not ready to do — I need get more comfortable with Python, and much more familiar with what Remix can and can’t do, before I can really get into that territory. (If you missed the previous post: afromb.py is a Python script that uses the Echo Nest Remix API, a music analysis and manipulation library, to rebuild a song (song “a”) from the pieces of another song (song “b”) by slicing both up into tiny bits and then trying to match those bits together heuristically. To put it simply. It’s the clever bit of guts at the center of these experiments. I did not write it, I’m just enjoying using it and trying to learn more about how the whole thing works.)

One of the common suggestions that is practical at this point is to, instead of running afromb.py against an entire mix, run it against each individual solo track in a mix — guitar, bass, vocals, drums, etc. So, the guitar track gets imitated as best as possible by foreign audio; the bass does too, separately; and the vox; and the drums; and then all of that gets mixed together again at the end. The notion is that the sum of the parts may be a lot more listenable than a single pass would be.

And I’m happy to say that, having tried that out and produced the media below, it does improve things somewhat. There’s a steady beat — thanks largely to being able to rebuild percussion tracks in isolation — and the overall dynamics of the song are a lot better too (it gets loud, it gets quieter, there’s notable section shifts). The track is still, note, not something that really sounds in a casual sense like The Hand That Feeds, or necessarily something you’d put on at a party; the resemblance is still primarily rhythmic, and while the overall feel is more musical than the tracks on Seattleward there still no real harmonic or melodic correspondence between the original and new creation. Solving that problem is for another day.

Building the track

Part of the challenge with this method (aside from it being a lot more time-consuming in general to produce a track) is getting my hands on those isolated instrumental and vocal tracks in the first place. Hence the return to Nine Inch Nails as source audio; they make available stems at remix.nin.com, which is honestly a fantastic move that I’d love to see more bands make. I grabbed the Garageband session for The Hand That Feeds (no Downward Spiral tracks available, and I’m out of date on NIN since like 1996, but I remember hearing this single when it came out), bounced out each of the seventeen individual tracks that make up the mix, and proceeded to throw the same Frasier clips at each of those as I did at the mixed tracks on Seattleward, using afromb.py to rebuild them.

Then I took those newly generated frasierplexes and imported them into the Garageband session alongside their respective source tracks from Trent and company, and made a rough mix of the new audio to try and generally mirror the sonic profile of the original. I threw a little distortion on the tracks that were replicating guitar and synth and bass, compressed things fairly aggressively, applied some panning to give it a little more sense of space and separation (if Frasier is saying four different things on four different tracks I want to at least give that some stereo spread to keep it from becoming totally mushy), put a little reverb and echo on a couple tracks, and, bam: The Crane That Feeds.

Producing the video was a cinch; I just used the vafromb.py script with the final mix as my a file and a random clip of Frasier singing on a telethon for the b file. Hence the weirdly camera-centric music video.

So what started yesterday as a silly photoshop joke turned, today, into a silly computational music generation sort-of-joke.

The Seattleward Spiral is a from-scratch, mostly-automatic reinterpretation of Trent et al’s 1994 The Downward Spiral, using nothing but sliced up audio from the NBC sitcom Frasier. The result is something arguably far more “industrial” in some sense (and far, far less listenable in that same sense) than the original album.

The reconstitution was done using “afromb.py”, a python script that takes two files, the target track a that you’re trying to recreate, and the source track b that provides the audio content used for that recreation, and slices both up into very small pieces, assembling the bits of track b one at a time according to whichever piece best matches the current bit of track a.

The script is provided with Echo Nest Remix, itself a fascinating and extensible music manipulation library and API I’ve been playing with lately and will have more to post about soon. Fun, fun stuff. If you’re not afraid of a little light monkeying around with your command line it’s worth checking out.

In any case: The Seattleward Spiral was created by running afromb against each track from Downward Spiral three times, using three different bits of Frasier audio — the theme song, and the first five minutes or so form two random episodes I found on Youtube (“Two Mrs. Cranes” and “Niles Starts a Fire” or something like that). I dumped the three tracks for each into Garageband to create a very basic mix with just some static panning to create a bit more stereo action, chucked those out to mp3, and that was that. It’s a tremendously quick, dirty mix, with functionally no human intervention in the choice and arrangement of the sounds on the record.

You can grab the whole thing via this zip file, or listen to individual tracks below. I think that Closer To Roz, in particular, works well.

Download audio file (josh-millard–nine-inch-niles–01-dr-self-diagnose.mp3)

Download audio file (josh-millard–nine-inch-niles–02-yuppie.mp3)
Download audio file (josh-millard–nine-inch-niles–03-marisy.mp3)
Download audio file (josh-millard–nine-inch-niles–04-march-of-the-prigs.mp3)
Download audio file (josh-millard–nine-inch-niles–05-closer-to-roz.mp3)
Download audio file (josh-millard–nine-inch-niles–06-frasier.mp3)
Download audio file (josh-millard–nine-inch-niles–07-the-becraning.mp3)
Download audio file (josh-millard–nine-inch-niles–08-i-will-not-drink-this.mp3)
Download audio file (josh-millard–nine-inch-niles–09-big-man-with-a-mic.mp3)
Download audio file (josh-millard–nine-inch-niles–10-a-rent-controlled-place.mp3)
Download audio file (josh-millard–nine-inch-niles–11-erasmus.mp3)
Download audio file (josh-millard–nine-inch-niles–12-jack-russell-terrier.mp3)
Download audio file (josh-millard–nine-inch-niles–13-the-seattleward-spiral.mp3)
Download audio file (josh-millard–nine-inch-niles–14-yurt.mp3)

UPDATE:

See also the Closer To Roz (Niles Mix) video created this evening using the related vafromb.py script! Note that the audio content of the remix is different here than what’s in the track above, as the input video is different; I also went ahead and kept a touch of the original track in there to make the video a little more coherent to listen to.

There were a couple challenges there, for me:

1. Instrumentation — what are the pieces of the “band” that need to end up on tape? With what mix of sounds does it stop sounding like a guy in his bedroom and start sounding like four or five people putting on a show?

2. Execution — what does it take to put those pieces together convincingly? How do you, mechanically, get from one guy’s ideas to a convincing full-band sound?

I want to talk mostly about the second point here — and that’s where the title of the blog post comes in — but I’ll cover the instrumentation thing first.

Rock bands have drums. Or: rock bands have rhythm sections. Something is underneath it all, driving it along. Drums and bass, working together to keep your shit in gear. Rhythm guitar adding to that, where applicable. These are the guys in your band who, if they’re doing their jobs right, are keeping things tight in the pocket and letting the vocals and the lead guitar lines and any other melodic elements move around freely without sounding stranded or unmoored or flighty.

A solo singer/songwriter/guitarist’s performance is going to be fundamentally different than a rock band’s performance, regardless of questions of skill or talent, because of this different musical dynamic. I’ve done a lot more of the former than the latter with my own work, and at times I very much miss being in a band and having that different dynamic available. Short of actually starting/joining a new band (and all the logistical and emotional complexities that could entail), my best bet is to learn to do it all myself. And I’ve been steadily chipping away at that problem over time.

So I got myself a drum kit last July, and have been having a good time with it and am learning to play them (relatively quickly, I guess, based on the feedback I’ve gotten from a few folks). A lot of years of musicianship beforehand have probably been invaluable on that front. Playing a little bit of Rock Band drums honestly didn’t hurt either. So going into this project, I knew I had drums, and I knew I could beat out at least basic tracks with some kind of steadiness.

We also own an electric bass, a fantastic little John Deere custom job with decent action and not-great pickups. I’m not much of a bassist but, again, I can knock out simple things — guitar skills transfer in a useful way there, though there are different things that work on guitar vs. bass and different physical approaches to the two instruments, and my bass fundamentals could use an awful lot of work.

Guitar I’ve played for years and years, so as a portion of the rhythm section that was a given, though I’m honestly not all that satisfied with my rhythm guitar chops. Familiarity breeds contempt, perhaps; remembering my early days of really questionable rhythm execution may come into it to, with me viewing my current skills more harshly than I ought to. It’s hard to say. But I don’t have any reason to be embarrassed about my playing in general.

Piano isn’t so much a standard rock component, but it’s hardly unprecedented and it’s definitely a viable component of a rhythm section. I’ve had a chance to put that into action in the past when playing with The Man So Cool, and had a lot of fun with it, and I own a nice electric piano that I’ve been playing a lot of for the last couple years. Both the left and the right hand work on a piano can tie into the beat in a way that helps reinforce the underlying rhythmic skeleton of a rock song.

So that’s a good set of sounds. Throw ‘em all in a basement and you’ve got a rock band. Right?

One nice thing about being in a band is that you, personally, generally only have to worry about one or two things at any given moment in a performance. Play your guitar, or your drums, or your bass, or your piano, or your triangle or theremin or musical bicycle. Maybe sing as well. Maybe shake your ass around a little too but that’s just icing. You’re paying attention to the other guys who are playing their own parts, but all you need to do is know that they’re doing their own thing and you’re all generally aware of each other, and, pow: a performance happens.

It’s one of the wonderful things about being in a band: something happens around you that’s bigger and more complicated than just the thing you’re doing yourself.

So there’s two big frustrations that come with building a fake band yourself out of overdubbed parts:

A. You have to do all the work.

B. You can’t jam.

Doing all the work means knowing how to play the instruments, as the first step. But beyond that, it means having to know how to make those instruments work together. It means having to be able to plan out an arrangement in four or five or however many parts and know what the drums are doing at the same time as the bass and the guitar and the piano and that so on are doing their own things, and anticipating the way in which those various parts will interact in a pleasing or exciting or surprising way.

It means, basically, being able to hear a whole rock song in your head in however many separate simultaneous pieces, and to be able to turn that into recordings, one track at a time, in a way that fits together cohesively in the end. Which is challenging in a lot of ways. Fun and rewarding if you pull it off well, certainly, or I wouldn’t try. But that’s the scope of the problem.

Not being able to jam, though — that’s a killer. And it’s a wall I hadn’t really run into before in the way that I did when I started working earnestly on the recordings for Inchoatery.

Obviously you can’t play everything at once, and obviously that creates a more significant lower limit on laying down a multi-part track — recording four or eight or twelve different instrumental parts one after another takes more time than recording a single live take (setting aside the question of how good any given take is and how many times you might need to try to get it right). That much I knew going in, and I generally work pretty fast when I’m recording a song so it wasn’t really a concern.

But there are other implications that I didn’t consider.

For one thing, the difference between recording one song and recording eleven songs brings into play the question of how you put down the individual instrumental tracks for each song. If I need to record drums, bass, a couple of guitars, piano, and vox for pretty much every track on the album, do I do that set of instruments for song one, and then move on to song two and record those again, and so on? Or do I record the drums for everything first, and then the bass for everything, and then the guitars, and so on? Neither is inherently correct (or incorrect). But!

I don’t have a recording loft with everything set up and ready to record at a moment’s notice; I have a house with instruments in various places, and a Macbook and a small audio interface and a small collection of microphones. If I want to record drums, I have to take my little studio to the drums and do a little setup before I can get started. If I want to record electric bass, I steal my kick drum mic and haul my gear over to the other end of the basement and set the mic up and tweak it and dial in an appropriate effects chain on my PodXT. And so on.

So the practical answer was to do all the drum takes at once. And then all the bass takes. And so on. Even granting that I’d probably come back to individual songs to retake something if the original takes didn’t work, I was motivated by the short timespan of the month and the desire to not lose my fucking mind setting up and tearing down a recording setup sixty six times instead of six times to take this assembly-line approach. To say nothing of the risk each time of setting things up inconsistent with a previous take and so having things sound uneven within a track or between album tracks.

The downside of that process is that it puts a big buffer, both in terms of time elapsing and in terms of the amount of musical data I’m shuffling around in my brain, between e.g. the point where I record a drum track and the point where I put down the bass track that goes with it. Which means that my memory of exactly what I did on the drums — specific kick patterns that I might want the bass to match, specific dynamic moves or sudden drops, small variations that I may have improvised while tracking the drums from one take to another — isn’t fresh in my mind when I go to add the next track. Some of that I can compensate for by reviewing the existing recording, or making explicit notes, or doing a lot of dry runs or repeated recording takes on the bass part to work out the details — but time was a factor here, and so was energy level. I couldn’t get this stuff perfect without killing myself in the process and running out the clock as well.

But it gets worse! The assembly-line approach introduces difficulty in tying any two tracks on a song together as tightly as I might like, but that’s true for every subsequent instrumental track as well. Small problems between drums and bass; small problems between guitar and the drum-and-bass undercoating; small problems between the piano and all the other stuff. Little issues stack up in a multiplicative fashion. Errors and mismatches get amplified.

And, the kicker: there’s no real time for do-overs if it all goes badly. Because of the time constraint of doing the whole thing in a month—which is practically reduced to recording the whole thing in a lot less than a month, since there’s a lot of writing and arranging and revising and planning that needs to happen before I can really seriously roll tape on the final tracks—it becomes a get-it-right-the-first-time situation. Or, at best, get-it-right-the-second time, if I’ve got the time and the energy to try.

So here’s the thing. With a band, if you’re all collaborating to put together new music from scratch, it often goes something like this:

ALICE AND THE ITERATORS
an algorithmic play

ACT I: The Idea

Alice: I have an idea for a song.  It goes something like this.  
Alice: [plays song idea]
Bob, Corin, Dio: Huh, cool.  Let's try it.
Alice, Bob, Corin, Dio: [plays basic parts around song idea a few times]

ACT II: The Refinement

Bob: Hey, what if we do this thing at this point in the song, like this?
Bob: [mimes or sings or plays or draws picture in dirt on basement floor 
    to convey musical idea]
Alice, Corin, Dio: Oh, you mean like this?
A, B, C, D: [tries out idea]
Corin: Wait, what if it was a little more like this?  
Corin: [Miming/signing/playing/scribbling commences]
A, B, C, D: [tries out idea]
Dio: Holy diver you've been down too long!
Bob: What he said.

ACT III - XXVI: Iteration

Alice: Hey, this is just Act II again a bunch of times but with different 
     specific ideas.
Corin: You write out the plans, I carry out the act.
Bob: What she said.  Oh, and what if at the end of the second verse we...

ACT XXVII - Satisfaction

Alice: Hey, after iterating through that song a number of times, we've 
    created a pretty solid finished product.  What a joy this process of 
    organic collaborative development can be!
Alice, Bob, Corin, Dio: [go to bar]

~FIN~

The missing piece of the puzzle, trying to do this all by myself, is Acts Three through Twenty-six. That process of steadily iterating through the “idea, implementation, evaluation, refinement” phase of songwriting is invaluable, and every good functioning band does it whether they think of it in such absolute terms or just let it happen.

I think of it in those terms, and think of the word “iteration” specifically, in no small part because of my formal Computer Science background—anyone who has been exposed to software engineering concepts will recognize the idea and the terminology immediately. (A related idea from CS is the “waterfall model”, which to the best of my knowledge has nothing to do with TLC.)

And trying to recreate this process of band-wise song iteration on my own — and failing to do so satisfactorily — was the biggest unanticipated roadblock I encountered when working on Inchoatery.

All in all, my challenges came from this combination of:

- not being able to — or not having the backing musicians around to help me — do everything at once (which, as a habitual solo recordist, I’m accustomed to), and
- wanting to really sell a full-rock-band sound (something I’m less used to), and
- trying to pull of nearly a dozen tracks at once (which I’m manifestly [HA!] unaccustomed to, as I usually work on one track at a time as they come to me throughout the year), and
- not having a group to collaborate with about changes and refinements (which I’m used to for solo work but not used to when developing full-on rock songs), and
- not having the time to, even however laboriously, create many iterations of the song on my own, because I had only a month to complete this project.

The result was that I was stuck with only my own ideas to work with, I was responsible for executing all of them as cleanly as possible (which was not in a lot of cases as cleanly as I’d like), and I didn’t even have the time to really tear down and rebuild those ideas and that execution enough to get the bugs out of the arrangements the way I’d be able to do if I was working with real live fellow musicians at a natural pace.

On the bright side, there were far fewer band arguments.

This isn’t intended as an earnest complaint: I chose to do this project in the first place because I’m attracted to the process and to the challenge, and pulling the album off as best as I can (and I feel I did a pretty damned good job, all in all) is a big part of the appeal of trying to do an album in a month in the first place.

I had a lot of fun doing this, and I’m really happy with the album that Inchoatery turned out to be.

A lot of what surprised me in the process of recording the album has been tremendously useful to me, too; every frustration has been educational, every roadblock a wonderful kick in the ass. I can too easily get used to riding on the easy paths of writing and recording to which I have become accustomed over the years, and that means it’s easy for me to stop learning, and a project like this is a great way to get out of that didactic slump. Certainly I’m better equipped to plan and execute a project like this in the future, thanks to all the messes I had to muddle through this time.

And there are solutions to a lot of the problems I ran into:

- Take more time. That’s not so much applicable to a stunt like album-in-a-month, but setting aside that arbitrary constraint there’s no reason I couldn’t have done more iterations of these songs and gotten more of the details pinned down, gotten more polish on the final recorded tracks, etc. Being able to just take a week or a month off from a song to let it sit and gel would eliminate a lot of the stress this project involved.

- Put it on paper, work it on paper. I have notes I made throughout the month, and basic sketch demos I recorded when an idea was starting to come together. I wish I had written more down instead of trying to carry ideas around in my head, and I wish I’d gone back and listened to those various demos a bit more to remind myself about some of the ideas I had that got left by the wayside throughout the month. If you think some specific thing might improve a track, write that down. It may save your ass later on when you review your notes and remember that great idea that slipped your mind.

- Seek help. I like that this was a 100% DIY effort, but the difference between doing it all myself and doing very nearly all of it myself is a point of silly pride, not a practical accomplishment. Get someone you know and trust to listen to what you’re working on early on, and throughout the project. Get more than one set of ears on it. Share your doubts; talk through your ambitious ideas; find out if there are problems you didn’t consider. Take advantage of whatever creative support network you might have to inject perspective into your creative process.

- Demo early, demo often. I created very rough sketch demos for most of the songs on Inchoatery, and decent full-band demos as well, and each time I did that I was forced to test my vague ideas and prove they worked. I could have done that more often before starting in on the final recordings and saved myself some pain. Every time you have an idea and you’re not absolutely sure how it’s going to work in reality, just try to record it. You’ll know immediately if you’ve actually got a plan or just a vague intention for how to execute that idea, and you can refine or abandon it from there, with confidence that that particular problem is taken care of.

Those are all lessons I’ll be making use of going forward, at least if I’m smart enough to remember them. If nothing else, I hope I come back to this and give it a re-read before next February rolls around.

Aside from providing me with an excuse to do some creative work and producing thereby a pretty solid, listenable album, this last month has been a deeply worthwhile investment in my own writing and producing skills. If you’ve thought about trying something like this but never quite made the leap, I heartily encourage you to go for it.

That page has the album available for immediate streaming, plus notes about the album in general and links to more details about (and various demo and sketch versions of) each of the songs on the album.

If you’d like to snag the whole album directly in mp3 form, you can download this zip file, as well.

I’m pretty happy with what I came up with, and it was a challenging and exciting and very educational month. I’ll be posting more details about the work I did and what I learned, but for now I’m just pleased to be able to say it’s finished and ready for consumption.