Let’s transcribe Yoshi’s Island!

A step-by-step guide

If transcription is an arduous hike, analyzing that transcription is the breathtaking view at the top
of the mountain.

There’s nothing better than transcribing a piece and diving inside the mind of the composer,
plumbing the depths of what makes it great.

It feels like digging for buried treasure and is almost as enjoyable as playing the music on an
instrument.

Here’s the step-by-step process we’ll be going over as we transcribe and analyze a piece
together:

 Stating your goals and hypothesis

 Preparing your transcription
 Transcribing a lead sheet

 Reducing a lead sheet (optional)

 Naming the chords

 Analyzing your finished transcription

 Remaking the piece (optional)

Stating your goals and hypothesis

I’ve found that going into a transcription session with a distinct goal in mind is the best way to
get the most out of the entire process. What do I mean by “goal”?

Well, as I alluded to in the action items of the last post, it’s important to ask yourself why you
want to transcribe the piece you’ve chosen. Obviously, there are objective reasons other than “it
sounds cool.”

I’ll tell you why I want to transcribe the Yoshi’s Island Map Theme. Well, for one, it’s short and
serves as a succinct case study for transcription.

But if I were to dive deeper, I’d say my goal is to figure out how the underlying harmony of the
piece (the chord changes, really) support the aesthetic of Yoshi’s Island. The game is hand-
drawn, childlike, and upbeat.

Yet, it still retains that signature Koji Kondo groove that’s present in all the previous Mario
titles.
My hypothesis is that this has something to do with the playful instrumentation. Remember the
scientific method from elementary school?

Like any good scientist, our goal is to conduct experiments that either prove or disprove our
hypothesis.

As you listen to your piece a few times, jot down these thoughts on a piece of paper or empty
text document. With our goal and hypothesis recorded, we’re ready to prepare our transcription!

Preparing Your Transcription

Before you dive into your notation program of choice, there are a few things you can do to make
your life easier as a transcriber. I recommend figuring out three key pieces of musical data before
transcribing:

Time signature

It’s important to understand your time signature before you start transcribing, especially when
dealing with odd meters like 3/4 or 6/8 (or even weirder stuff than that).

Most video game music is in 4/4 time, but it’s important to listen to the song, count the beats in
each measure and make a mental note of this going into your transcription process.
What I like to do is tap out what I believe the time signature is with my foot (in the Map Theme
it’s 4/4), and then tap out the melody with my hand on my desk.

This takes a bit of practice, but I’ve successfully confirmed the song to be in 4/4 time. As a side
note, this tapping exercise has also revealed that the melody is heavily syncopated (emphases are
placed on the weak beats or in between beats).

I’ll keep this in mind for the analysis phase.

Key signature

If you can figure out what key your song is in before you start transcribing, it will save you a ton
of grief and potentially spare you from dealing with unnecessary accidentals (sharps and flats).

Finding the tonic (key center) takes a little bit of ear training, but most often, it’s the first chord
in the form. Playing along with the song on a piano or guitar is the best way to figure out this
out.

I wish I could help more here, but the best I can offer is when you play or sing a note that feels
like “home,” that is likely the tonic and therefore the key center.

If you’re having trouble, there are a few other ways to figure out the tonic, as illustrated by
talented saxophonist and YouTuber Jeff Schneider in this video.

Tempo and form

If you’re using software, you obviously want the tempo of your transcription project to be as
similar as possible to the tempo of the song.

I love this Tap Tempo tool, which not only allows me to tap in tempos but also keeps track of my
beats (this will come in handy soon).

The first thing I usually do is pull up my song of choice in a separate window on YouTube or
TuneTranscriber.

This next part requires a bit of keyboard and mouse finesse. Put your mouse cursor into the tap
tempo text box on the BPM counter tool, then click on the YouTube video to begin playback.

Quickly click back to the BPM counter and wait for the form to loop (most YouTube videos
have looping versions of these songs).

As the song loops back to the beginning, I begin tapping my spacebar to the beat, until the song
reaches completion or the loop point once again.

If you miss a few beat taps here or there, it’s not a big deal. As long as you get the estimated
amount of beat taps, you’ll be set to start outlining the form in your notation software.
Here are my results from the Yoshi’s Island Map Theme:

Using this method, I tapped out 18 beats in the form, with an average tempo of 121 BPM (as
seen in the screen capture).

Now, if you followed the above steps, you should know your time signature, which in the case of
this piece, is 4/4.

Logically, if there are four beats in a measure, we can divide our total number of beats by 4 and
the result will be how many measures we need to create for the form. In this case, 18/4 = 4.5.

That .5 measure likely accounts for one measure of 2/4 that acts as a turnaround for the loop.

So now, when MuseScore or Sibelius prompts us to give some rudimentary information about
our score, we’re set up for success knowing:

 The time signature is 4/4

 The key center is G (one sharp)

 The BPM is 121

 There are 4 measures of 4/4 and one measure of 2/4 at the end

It drastically speeds up your workflow to have a clear blueprint of the song from start to finish.
This saves your creative energy for the hard work of transcription and just feels great.
Transcribing a lead sheet

A lead sheet is essentially a piece of music reduced to its core musical elements.

This means it contains only the most basic rhythmic, harmonic, and melodic content needed to
communicate the “essence” of the piece.

This is a perfect place to start for transcribers, and Gordy Haab mentioned this is how he started
transcribing the Star Wars score for his work on the Battlefront games.

It’s easy to get lost in the weeds transcribing arpeggios and inner voices. What we want is the
core musical data of the piece, and then we can decide if we want to go further from there.

During our interview, Gordy outlined a very helpful process for creating a lead sheet.

Let’s follow his process together as I transcribe the Map Theme from Yoshi’s Island.

Check out the Transcription and Analysis Cheat Sheet for a quick reference you can have by
your side every time you transcribe.

First, transcribe the melody

9 times out of 10, the most obvious thing you’ll hear in a piece is the melody. Transcribing the
melody first is a great way to warm up your ears for the more challenging inner voices.

Use TuneTranscriber or YouTube to play the song (utilizing the slow-down function, if need be),
and transcribe the melody.

Starting out, I recommend taking it beat-by-beat, not measure-by-measure. Slow down the piece
until you can count out “one – ee – and – ah -” at a comfortable pace.

“One” marks the beginning of the beat, “and” marks the middle point of the beat (eighth note),
and the “ee” and “ah” mark the 16th note subdivisions of the beat.

I’m going to use this practice to count out the first beat of the Yoshi’s Island theme. First, I’ll
slow it down, and then count it out. I notice when I count it out that there are notes on the “one –
ee – and” but not the final 16th note “ah.”

“Boppa Ba Ba!”

I’d notate that like this:

The second measure, counting nice and slow, hits the melody on these beats:

“Two – ee – and – ah”

This straight eighth note pair provides a nice contrast from the syncopation in both the previous
and subsequent measure.

I’d notate it like this:

If you need a refresher on how to write musical notation properly, I recommend this helpful
video by 12tone.

Going beat-by-beat, slowly but surely, I end up with this transcribed melody:

Second, transcribe the harmony

The next step in Gordy’s process is to transcribe the chords.

For old systems like the SNES, NES, or Genesis, chords can often be voiced in unusual or
especially funky ways due to the limitations of the hardware.

Because of this, it’s often a good idea to transcribe each voice separately, starting with the bass
note.

This kind of goes beyond the bounds of a lead sheet, but if you’re not sure what the chords are, it
can be helpful to get everything down “on paper” and analyze the harmony later.

Here’s the Map Theme with the bass transcribed. Note how there are no chords labels yet
because many of the inner voices inform what chords we’ll be naming.

Third, transcribe the inner voices

The last step would be transcribing some of those hard-to-hear inner voices, which should be a
bit easier now that you have the top boundary (the melody) and the bottom boundary (the
bassline) transcribed.

It’s especially helpful to use an instrument as you figure out these pitches, and don’t be afraid to
reference Game Music Appreciation (for retro tunes) to isolate the channels to double-check
your accuracy.

This is the part of transcription that will really train your inner ear and push you to your musical
limits. It’s hard work!

Here’s the full Map Theme with the inner voices transcribed. Full disclosure, I referenced Game
Music Appreciation to double-check my work. No shame!

Sidenote: Reducing to a Lead Sheet

It’s important to note here that if you’re transcribing an orchestral piece for the purpose of
musical analysis, there’s an extra step between the transcription and analysis phases called
reduction.

Let’s say you want to transcribe the Main Title Theme from Bioshock. You meticulously listen
to each section of the orchestra and transcribe as you go.

The next thing you know, you have 16+ staves with more notes than your brain can comprehend
at once.
Don’t be a hero and try to analyze this as-is. Reduce it!
(This also applies to VGM pieces that include lots of complex parts, voices, or otherwise
convoluted arrangement choices.)

How in the world can you analyze something like this? Answer: reduce it.

I’m pulling this straight from Wikipedia:

“In music, a reduction is an arrangement or transcription of an existing score or composition in

which complexity is lessened to make analysis, performance, or practice easier or clearer; the
number of parts may be reduced or rhythm may be simplified, such as through the use of block
chords.”

When Gordy Haab was charged to write music for the new Star Wars Battlefront series, he knew
he needed to dive deep into the mind of John Williams through diligent transcription and
reduction.

What he already knew, and what you’ll find if you practice this craft, is that many sections of an
orchestra are playing the same thing. Sometimes in octaves, sometimes in unison.

“[Reduction] is like stripping a house down to its blueprint,” he said. “It’s the best way to truly
understand how an orchestral piece functions.”

Trying to analyze a full orchestral score, especially when dealing with instruments like the
trumpet and french horn that are transposed into other keys, can be a real headache without
reduction.
Plus, there’s no better way to hone your orchestral arrangement chops than by reducing
orchestral pieces and studying how they work.

Here’s Gordy’s process for reduction, taken straight from our interview:

 Start by combining all the winds (the first section in typical score order)

 Reduce each section sequentially (percussion is mostly just for color, but take note of
which parts percussion is accenting)

 Continue until you reduce the score down to one piano grand staff, maybe two at the
most

 Sometimes ornamentations like woodwind runs will disappear. That’s okay, you’re trying
to capture the essential harmonic and melodic information of the piece

“I had assignments in university where I had to reduce scores down to the piano. You learn a ton
about orchestration that way,” Gordy said.

“For example, I learned in John Williams’ music that the 4th trumpet would often double french
horns playing in unison. This subtle shift in timbre is part of what makes Williams’ music great.
I would’ve never noticed that if I hadn’t reduced it down to the simple ingredients of what’s
being played regardless of the instruments playing them.”
He also recommended a clever exercise for composers looking to build take their arrangement
skills to the next level.

“One thing I’ll do is actually break it back out to full orchestra after I’ve reduced it. This time,
I’m orchestrating it my way without regard for the original. Then, I compare the two. It’s just a
great exercise to hone your craft.”

Okay, back to our Map Theme!

Identifying and Naming Chord Tones

Now that we’ve got our raw musical data on paper, it’s time to name some chords. Why bother
naming chords, you ask?

Without chords, it’s a challenge to discern harmonic patterns.

Some chord names are obvious, but I found myself majorly tripped up by this step when I first
started transcribing.

Which pitches make up the chord and which are just part of the melody? How do I know if a bar
needs one chord, two chords, or more? How can I “feel” the changes and notate them
accordingly?

8-bit Music Theory analyzes video game music for a living. As such, he’s developed a solid
chord-naming process.

You’ll soon develop your own process as you transcribe more and more, but for now, let’s
follow Mr. Theory as a starting point (remember how much I love copying?).

Which pitches make up the harmony of the bar?

One of the biggest challenges in the harmonic analysis of transcription is deciphering which
notes in a composition make up the underlying harmony, and which notes are melodic
embellishments.

One of the helpful ways to crack this code is to organize notes in “stacks of thirds,” or logical
groupings that help better illuminate their function.

For a brief refresher, a third is an interval (two notes) three letter names apart. For example, A to
C is a minor 3rd, F# to A# is a major 3rd, and D♭ to F is a major 3rd.

The reason I know each of these are either minor or major is a mixture of scale knowledge (what
makes up the A, F#, and D♭ minor and major scales), but also how minor and major thirds are
formed.
The formula for major thirds is whole step – whole step. Meaning you move one whole step from
D to E♭, then one more whole step from E♭ to F. That’s why D♭ to F is a major 3rd.

Likewise, the formula for minor 3rds is whole step – half step. Which is why in the case of A to
C, we move one whole step from A to B, then one half step from B to C. There’s our minor 3rd!

Sidenote: If all of these sounds completely foreign to you, I’d recommend brushing up on at least
the basics of chord construction here.

8-bit graciously opened the kimono on his process for deciding how to prioritize certain notes as
chord tones. Here’s what he told me:

 Look at a measure and see if a chord jumps out at you (sometimes it’s obvious)

 If it’s not obvious, take all of the notes in the bar and see if you can arrange them in a
way that makes either a scale or chord. Sometimes it can be helpful to write out the bass
note as “1,” and the subsequent notes as scale degrees relating back to that note. (C, F#,
G, D would be written as 1, #4, 5, 9.)

 After rearranging these notes, look for key intervals that define chord quality like 3rds or
7ths. If you can’t find any of those, try another pitch as the bass note.

 Still no luck? Take only the notes that are emphasized (usually found on the beat, but not
always) and repeat steps 2 and 3 with just these notes

 If this still doesn’t work, isolate just the notes found on beats 1 and 3 (strong beats) and
repeat step 2 and 1 with just these notes

 If this somehow still doesn’t work, then you probably need two chord symbols per bar

Following this process will help you logically group together pitches. Some pieces demand one
chord symbol per bar, others demand two.
The Map Theme actually feels best with four chord changes, as we’ll explore soon. A lot of this
comes with experience, practice, and “feeling” the changes.

Sidenote: for those of you who play guitar or piano, I often think to myself as I’m transcribing,
“if I were performing this song live, where would it feel appropriate to change chords?”

It’s kind of a subjective method, but this has gotten me out of many tricky chord-naming
situations.

Want a handy guide as you transcribe and analyze your pieces? I made this Transcription,
Reduction, and Analysis Cheat Sheet so you can have a step-by-step checklist the next time you
transcribe!

Chord Naming Process

Once you understand how you might group your pitches together, it’s time to name the chords
themselves.

The priority for naming chords is usually as follows (extrapolated from this site):

 Start with the root of the chord

 If the third degree is minor, write m next to it, otherwise, write nothing

 If the seventh degree is minor, put 7. If it is major put maj7. Otherwise, write nothing

 If there is no 3rd degree, look for a 2 or a 4, and write sus4 or sus2 after the 7

 If the fifth degree is perfect, write nothing. If it is diminished, write♭5, if it is augmented

write ♯5. Sometimes a chord may have both♭5 and ♯5.

 Next, the ninth degree. If it is perfect, erase the 7 and write a 9 in its place. If it is minor,
write♭9. If no 7 is included, write add9.

 Next, the eleventh degree. If it is perfect, erase the 7 or 9 and write an 11 in its place. If it
is augmented, put ♯11. If no 7 is included, write add11.

 Next, the thirteenth degree. If it is perfect, erase the 7, 9, 0r 11 and write a 13 in its place.
If it is minor, put♭13. If no 7 is included, write add13.

Using this method, we have successfully identified and named this chord as Am7! Nice!
However, not all chords are going to be this simple. For example, take the chord in this bar.

Let’s follow 8-bit’s process and see what we find.

When we spell out the pitches, here’s what we get (along with their corresponding scale
degrees):

D, E, C, B

1, 2, 7, 13

Now let’s look for 3rds, and then 7ths. Hmm, so there aren’t any 3rds, and there’s a 7th in there,
so our initial impulse might be to label this as a D7 chord.

But upon further examination, we also see that B in there, making it a D13. But then there’s also
no 3rd in there, so we need to make sure that’s reflected in the chord.

3rds are super important because they define the quality (either major or minor) of a chord. Right
now, this chord could technically be either, though it’s most likely major.

So we might end up with something like D13sus2, which is a very uncommon chord and likely
won’t help us much in our analysis phase.

Sometimes it’s good to back up and not necessarily start at the bass note, but split up the “right
and left hands” so to speak, and use slash chords.

If a chord is built on top of a bass note clearly distinct in quality from the rest of the chord, you
may choose to write a slash chord.

In the case of the above chord, the right hand is playing 2/3 of a Cmaj7 chord (the bold pitches),
and the bass is hitting a D. Let’s take another look:

D, (E, C, B)

1, 2, 7, 13

A Cmaj7 chord is made up of C, E, G, and B.

All of those pitch requirements are met in this chord (aside from the implied 5th, G, which we
don’t really need).

Because of this, my preference is to write this chord as Cmaj7/D.

This also simplifies things in a live musical setting where a musician or group of musicians can
separate the bass from the other voices of the chord. It makes at-a-glance analysis a lot easier, as
well.

Helpful chord identifying tools

I recommend that before you turn to any online tools, you put on your thinking cap and give the
above process a college try.

But like I said earlier, the goal is to have an accurate transcription. So if you’re absolutely
stumped or just want to check your work (I recommend it), punch your pitches into this Chord
Identifier tool.

When it spits out your answer, don’t just enter it into your transcription. Click on the chord and
study why that chord is named that way, and what went wrong with your initial assessment. It’s
the best way to grow as a music theorist and chord namer!

Some chords are open to interpretation

Don’t let anybody tell you there’s only one way to name chords. This is where the process of
musical analysis gets grey and subjective.

I’ll give you an example.

The other day I was transcribing the Muda Kingdom theme from Super Mario Land (great tune).
The first chord change sees the bass shifting from the I (A) to the IV (D), with the melody hitting
the B, C# and A.

On first blush, these pitches lined up (D, B, C#, A) might tempt me to label the chord as an
Dmaj13.

But when I played the cluster chord on the keyboard while singing the melody (a helpful
exercise), I felt like it sounded more natural as a Dmaj7. I resigned that B the melody hits to a
melodic embellishment, not part of the chord itself.

This is very much a “lead sheet” philosophy. If there are no altered tones (♭9, #11, etc.), that
fundamentally change the tonal color of the chord, it’s often best to name the chord a major 7 or
dominant 7. This leaves the embellishments up to the discretion of the player or the melody.
What do you think? You may disagree, as I imagine some people love to label exactly what the
pitches are implying (including embellishments). I tend to have a looser view of this, especially

because I’ve played with lots of bands and like to leave room for surprises.

Roman Numeral Analysis

My favorite part of analyzing music is conducting a Roman Numeral Analysis. If transcription is

the arduous hike up the mountain, analysis the view from the mountaintop, Roman Numeral
Analysis is the selfie you take to make all your friends jealous.

After I’m confident my chord symbols are as accurate as they’ll ever be, I print out my sheet
music and start marking things up with a pen.

Whatever key the song is in, I notate as a “one” chord. Except I don’t write “one,” I write “I” or
“i,” depending on if the song is in a major or minor key (uppercase for major, lowercase for
minor).

From there, every scale degree relates back to that I chord in some way. For example, if I’m back
in the key of C major, I’d label all Dm chords “ii,” all G7 chords “V7,” and all F major chords
“IV.”
Each Roman numeral relates back to how these chords function in relation to the tonic chord.

Musictheory.net has a thorough, concise tutorial on Roman Numeral Analysis here. I highly
recommend you check it out!

The reason why this is important is that I’m not so much concerned with the actual chords, but
their function. This is what Roman Numeral Analysis helps me discover.

A ii – V – I cadence has the same function in every key, and if you want to get at the underlying
skeleton of a song, this is the best way to do it.

I can’t overstate how important this is in learning genre-specific cliches and how they’re used in
real compositions, as well as the theory behind how tonal music works.

Using the circle of fifths

The circle of fifths and I have gone through a rocky relationship. From slight interest to mild
indifference to angry contention, to true love.

The Circle really does contain everything you need to know about harmonic analysis and more.
It’s like the golden ratio or yin-yang of music.
Everything in perfect balance. Everything in harmony and symmetry.

I could write a whole blog post about how to use it for analysis (and I will), but for now, here are
a few basic things to know as you reference this amazing tool in your analysis process:

1. The circle moves clockwise in fifths, and counter-clockwise in fourths (each chord
resolves to the chord in the counterclockwise direction)
2. The relative minor of each key is three steps to the right of a given chord (Am is three
steps from C)
3. Drawing a line straight across the circle produces a tritone (C to F#)
4. The chromatic mediant chord (often used in film scores) is three steps to the left of a
given chord (C to Eb)
5. Starting five steps to the right of a chord and sequentially moving counterclockwise back
to the key center will produce the very popular and effective viio – iii – vi – ii – V7 – I
chord progression (Bo, Em, Am, Dm, G7, C)
 6. Drawing an equilateral triangle starting at the key center produces a Coltrane Matrix
(chromatic third relations), a super spicy jazz progression popularized by John Coltrane
himself

I could go on and on about how great of a tool this is for writing, how many creative blocks this
thing has rescued me from, and how I discover new things about it every day.

I’ll restrain myself and save that for another post.

I encourage you to start viewing the music you analyze through the lens of the Circle of Fifths.

I often reference it as I’m doing a Roman Numeral Analysis to make sure I’m accurate.

After the analysis, I’ll look through the circle of fifths to see what patterns the song follows.

For example, when I analyzed the Menu Themes from the NES and SNES Classic, I made a
video about how they traversed the circle of fifths in super unique ways (and even composed a
piece in the same style).

A few more tips for analysis

In addition to studying the piece’s harmonic language and chord progression, there are many
other ways to analyze a transcribed piece of music.

After naming the chords, I usually follow this process:

 Study the “line” of each voice (this is usually called voice leading) for patterns

 Try to find motifs that recur in the music and figure out how they function (I usually look
for similar rhythms first, then similar pitches)

 Study the rhythms of the piece and what makes them work. The best video game
composers often utilize syncopation (placing emphasis on the weak beats) to drive a song
forward

Analyzing your finished transcription

So now that our piece is transcribed, our chords are named, and our Roman Numeral Analysis is
complete, it’s time to test our hypothesis against our findings.

If you’ll recall from many paragraphs back, my goal for this piece was to understand how it
supported the childlike aesthetic of Yoshi’s Island while still retaining that jazzy, catchy Mario
sound.

My hypothesis is that it was because of the instrumentation (percussive, plucky sounds).

When I examine the bassline of the piece, there is a feeling of rising up the scale, then falling
down the scale, then rising up again and falling down again with a little twist (I’ll get to that in a
second).

Many nursery rhymes have elements (either melodic or harmonic) that ascend the scale in
simple, small steps. This makes it easier for children to sing. This is why combined with my
theory on instrumentation, I believe it has a child-like sound.

But about that signature Mario catchiness? Remember how I said when I tapped out the tempo
and melody, I noticed the tune was extremely syncopated? This is a Koji Kondo staple, and
accenting rhythms in between beats or on weak beats is an effective way to drive a piece
forward.

It also has the benefit of making melodies really catchy. Is it just me, or when you play this
adorable tune, are you tempted to bob your head and tap your foot? That’s the syncopation
working its magic.

If we were to straighten out these notes, the song would sound dry, boring, and not near as
catchy. The straight, even bass notes act as the “straight man” against the wildly syncopated
melody in the right hand.

Koji doesn’t stop there, though. To put a little more stank on this tune, he borrows a B♭from the
parallel key of G minor, substituting it for the B natural we’ve grown accustomed to. The result
is a slightly-bluesy, delightfully unexpected shot of funkiness.
When I was processing this piece with 8-bit, he made the comment, “It’s a funky melody with a
baby-like nursery rhyme in the bass. Perfect for Baby Mario riding on Yoshi’s back.”

Boom. Analyzed.