How to Write a Melody: Finding Inspiration in Human Speech

Learn how to write a melody so that’s it more relatable able to connect with the listener, using three essential traits. Watch Joe Hanley, the creator of Syntorial and Building Blocks, as he walks you through the process of changing a robotic arpeggiation into an expressive melody. 

Try the full version of this lesson, which includes an interactive ear training challenge in our online DAW, as part of Building Blocks. Building Blocks is an online music production and composition training course that will teach you how to write music in a DAW.

Injecting different aspects of human speech into a melody makes it more relatable, and helps it connect to the listener. A great way to see exactly how this works is by starting with an arpeggiation:


An arpeggiation takes the notes of a chord and repeats them up and down in a pattern. They tend to be very repetitive and robotic, and thus lacking in human speech-like qualities. So how do we take this robotic, repetitive arpeggiation and make it more human, more expressive? How do we turn it into something that we can relate to and connect to?

We do this with three key elements: space, variation, and phrasing. Let’s talk about space first.


When we talk, we don’t constantly say one word after another with no space in between. For example, we put space between sentences. This is the first sentence…..and this is the second sentence. Sometimes we may put space in the middle of a sentence. Maybe I put it right…about there.

So, that’s the first thing we’ll add to our melody. Let’s take the arpeggiation and put some space in it:


This space allows our melody to breathe, and it gives our ears break from the constant pattern of notes. More importantly, it makes it more like human speech, and thus more relatable.

Now let’s look at trait number two, variation.


Now, are melody already has variation in its pitch. But its rhythm is very uniform and repetitive: one constant sixteenth note after another.

But when we talk, all of our words aren’t exactly the same length. Some are long, and some are short. Sometimes we talk quickly, and sometimes we talk slowly. Our words’ duration and rhythm vary. So, let’s start with the melody’s first phrase and change its rhythm by making some durations longer:

first phrase variation.png

Now let’s move onto the second phrase. We’re going to change its durations as well, but we also need to make sure its different from the first phrase. So we’ll start by moving the entire second phrase over:

second phrase shifted.png

And now we’ll change its durations as well:

second phrase variation.png

Notice how I also added some additional space within the second phrase. It’s important to allow yourself to drift between these various traits as ideas come to you.

With variation added in, this melody is really starting to sound like a true melody. The goal is to make it feel more like something that’s being sung. More like the way people talk. Which brings us to our third trait, phrasing.


Phrasing puts the finishing touch on a melody. Spacing and variation will get you most of the way there in creating a very human melody. But the phrasing is what really finishes it off. It’s a finer point of creating a great melody.

One very common example: the question and answer pattern of phrasing. When a person asks a question, they often end it on a higher pitch. For example, in the phrase “What is your name?” the word “name” is often spoken with a higher pitch than the previous words.

But when someone answers a question, like with the phrase “My name is Joe,” they often end it on a lower pitch. In this case, “Joe” would have a lower pitch than the previous words.

We can take this question-and-answer speech pattern and use it in our melody. For this melody, it’s as simple as raising the last pitch of the first phrase. So now, the first phrase is a question, and the second phrase is an answer:

q&a phrasing.png

This is yet another step towards making the melody mimic and human speech and thus making it more human overall. This gives it life, makes it more inspiring, and simply makes it nicer to listen to. As a result, when it comes to the sort of phrasing born from human speech patterns, there is an endless variety.

Here’s another example. Let’s say I list two statements in a row. I say, “this is the first statement.” Followed by, “And this is the second statement.” The first statement is often very plain. However, sometimes that second statement starts at a higher pitch and emphasizes the first word. “This is the first statement…And THIS is the second statement.”

So let’s create a brand new melody that mimics that phrasing. We’ll begin with a phrase that starts in the mid-range and then works its way down:

two statements first phrase.png

Then, for the second phrase, we’ll jump to a higher pitch and hang on that first note for a bit:

two statements both phrases.png
Open and play this example in the Studio

So how can you start finding different speech patterns to use in your melodies? Try recording yourself talking, and then listening back. Or record someone else talking. Or go online and watch an interview with someone.

There are endless ways you can start listening to how people speak. Specifically, focus on their pitch and rhythm. And see how variation and space play a part as well. Then let what you hear inspire a melody.