“Voice” Drum Transcription — Simon Phillips with Hiromi

Posted on August 20, 2023


Thus far, I’ve been slacking off on transcribing this year, so let’s get back into things with Simon Phillips. Today’s tune is “Voice” by Japanese pianist Hiromi — track 1 from the homonymous 2011 album. 

The band on Voice is a trio; Simon and Hiromi are joined by bassist Anthony Jackson, and this three–piece recorded four albums total. Voice was followed up by Move in 2012, Alive in 2014, and Spark in 2016. 

I discovered Voice via a random Apple Music suggestion, and it’s a pretty wild tune, with many time signature changes; it’s the kind of stuff you’d expect from something like prog or djent more so than jazz. While it’s a fun track, some of the playing does get a little too... bebop–ey for my tastes. I use that term very broadly to refer to the kind of jazz that’s stereotypically inaccessible to the average listener, especially because of dense harmonies. (You things are serious when you find sheet music that says “Don’t worry about the exact pitches.” More on that in a second.)

Before I start a transcription, I always look to see if someone has already transcribed what I want to look at. While I couldn’t find any drum scores, someone by the name of Han Zhao actually transcribed all nine minutes of the bass and the piano for “Voice”, which gives you an opportunity to analyze the tune as a whole:

After a free–form piano intro, the main riff starts up. In the above video, this riff is notated as alternating between 14/8 and 13/8.

This is a bit odd to me… at least, it was at first. Personally, my initial thought was to use 7/8, ending each phrase with a bar of 6/8:

There are a few other strange meter choices throughout Han’s score. There’s an ensemble stop time section that features bars of 6/4, 6/8, and 12/8 — my initial thought would be to just stick with 3/4. There’s another big unison thing written in 16/8, which is hard for me to make sense of; even with the syncopation, I don’t see a reason to not use 4/4. Moreover, I would have replaced 14/8 sections with either two bars of 7/8 or one bar of 7/4. 

14/8 is a strange time signature, and I struggle to come up with justifications for using it outside of some contorted beat partitioning (e.g. 2 + 3 + 3 + 2 + 2 + 2). Whether or not such a groove is best written in 14/8 versus two bars of smaller meters (e.g. 4/4 and 6/8) would be anyone’s guess, but I tend to not like lengthy time signatures.

I don’t want to beat up on Han’s score too much, because it’s very impressive even with my rhythmic nitpicks. 

After 2 ½ minutes of the trio jamming, we get the solos. First up is a long piano solo starting in 9/4, backed by this vamp:

I’m not kidding when I say long — it’s over 2 ½ in length just on its own! The solo eventually climaxes in “7”, whether you want to count it as 7/8, 7/4, or 14/8. 

The first time I heard “Voice”, I thought the solos were in 9/8. On the other hand, putting it in 9/4 gives Simon’s playing a half–time feel. Much like long time signatures, I try to avoid defaulting to half–time unless it’s the only practical way (or if I know for sure that’s how the performers counted it).

Cutting the tempo in half to give the tune a 2 & 4 feel would have the main riff alternating between 7/8 and 13/16:

Is that any better than half–time? Looking at the beams alone… probably not. 

If we do count the solos in 9/4, then it might make the most sense to count the “7” sections as 7/4. 

I’m beginning to see why Han used 14/8 and 13/8 for the head — alternating between 7/4 and 13/8 is quite gnarly, so 14/8 was used to maintain an 8th note pulse. 14/8 is still used even when there’s no alternating meter, perhaps for consistency. 

Because the head follows a classic 7/8 partitioning (2 + 2 + 3), I’m still drawn to using 7/8, but 9/4 feels like the best way to count to solo sections. As such, I’ve also used 7/4 when appropriate.  

Phew... we got a bit sidetracked there. Anyhoo, let’s get to why you’re even reading this. Next up in the tune is the focus of today’s transcription, a drum solo! Much like the piano solo, things start in 9/4 with the same vamp, before eventually getting to 7/4.

The drum solo starts with some grooving before moving into some triplets:

Look at that: over–the–barline triplets! If you didn’t know, there is a way to do this in MuseScore. A kludgy way that is.

Start by entering everything in those two measures except the triplet that will go over the barline:

Then, use the “join selected measures” tool to combine the two bars into one giant bar of 18/4 (note the time signature does not change, but a gray “+” will show up in MuseScore marking the measure):

As you can see, we have a half note rest sitting where our triplet needs to go. Turn the rest into a quarter note triplet (⌘ 3), and manually add in your barline from the pallets:

It’s just that easy folks. /s

Here’s another discussion: should over–the–barline tuplets even be used? It’s more common see normal subdivisions beamed over barlines, but this is mostly in older scores, which makes me wonder if it’s archaic. These days, stretching a tuplet over–the–barline is mostly a drum corps thing for dense tuplets that can’t be easily subdivided the way a quarter note triplet could. I mainly used it today for variety, and because I’ve always wanted to. 

The solo really starts to pick up with a flurry of 8th note triplets, featuring a deluge of syncopations. Things seem mostly ad–libbed, but there's a five–note motif that shows up a few times. You can see it in this excerpt, starting on beat 6 of the first line:

It was quite a challenge to keep track of the beat in a big measure of 9/4, so during my transcription process, I actually added in the vamp underneath the staff:

That way I could visually double–check what I was hearing on the track.

After the triplets comes an onslaught of 16th notes:

Note that these singles are played pretty softly, making them even more impressive at this speed.

It’s time to talk about Simon’s kit. To put it mildly... Simon likes big drum sets:

In that pic above, he must have at least seven toms, not counting octobans or that big “gong” drum on his right. Even more confusing, he has six different drum sets laid out on his website, and each one has at least seven toms in one configuration or another. 

I can’t even begin to figure out which one he used for this session, and whether or not he used his full get–up or a truncated kit. The only clue I can get comes from a 2015 post shared to the website formerly known as Twitter where Simon shows off his kit (a Star Walnut I believe) for a Hiromi gig in Mexico City, with four rack toms and three floor toms:

The toms on “Voice” most definitely blur together, and there comes a point where notating six rack toms starts to feel pointless. Moments like this convince me that the average listener probably just hears a high tom and a low one, save for moments where you just rip across the kit. MuseScore’s default MIDI drum kit only has six toms anyway, so I have four rack toms and two floor toms notated. God help me if he actually used that gong drum and I just missed it. 

Because the measures have an odd number of beats, there may be some surreptitious sticking changes amidst the singles. Simon also likes changing to open–handed playing on the fly; another thing to be aware of. 

As we move into the solo’s climax, there’s a cool slowdown effect using 8th note quintuplets, 8th notes, and then quarter note triplets:

The double kick starts to show up next, beginning with a quad fills, and then a triplet thing that’s probably played KKR (it might be KKL or a combination of both — there’s really no way to tell):

Note how the triplet double pedal lick morphs right into quarter note triplet hits.

The drum solo wraps up with a heavy groove before we go into the 7/4 section.

I wanted to look at Simon’s timekeeping on “Voice” — I was tempted to do the solo that comes just before the drum feature (as I did with “Get It Like That”), but the piano solo on this tune was just too long, so I look at the 7/4 jam after the drum solo.

I close out the score with the stop time section (this is the phrase Han notated with 14/8 and 13/8). I went ahead and notated it my way using 7/8 and 6/8. It’s kind of moot; I can’t notate it the way Han did because MuseScore still can’t do proper alternating time signatures since there are other priorities as of late. (Unlike over–the–barline tuplets, there are no kludgy workarounds to be found here, unless you have an insane amount of patience.)

Shortly afterward was the moment I realized the utility of switching back and forth between 3/4 and 6/8. The kicker (ba dum tss) was a big double bass fill:

While you could notate it in 3/4, it’s bigly awkward to do so. Both 3/4 and 6/8 have the same number of notes, but changing between them to communicate different feels isn’t that harebrained — a famous example is “America” from West Side Story.

“Voice” ends with another 7/4 jam, and then a hilariously syncopated stop–time section; have a look at the video to get the rhythms down. The finale is a sparse solo piano movement. 

So… this one was a doozy. I’m surprised I haven’t been hearing that vamp in my dreams yet.

Now would normally be the time for MuseScore 4 Bug of the Week. I don’t really have any new glitches to share (yet). I found several, but I can’t reproduce them reliably, which is no help. I will say that MuseScore did crash on me at least twenty times during this project. That has to be the most crashes I’ve experienced by at least twenty. Thankfully, autosave is working well, so I never had to redo more than a couple of measures. 

I would love to keep a log of every mouse click and keystroke I do in MuseScore to help reproduce crashes, except, I actually don’t want to do the developers’ job for them.

Download PDF.

Subscribe to the Blog!